I have purchased Metal 2 SHPN with mikrotik 5.18 Version , and after alot of settings , antennas changing all the time , poe changing , ehternet wire changing , i am getting the problem of response time of about 2000-3000ms and no data communication at all , poe cable length is 15meters and i have configured it on ( B) data rate of about 11mbps , all the settings are good it is new device my clients are tplink 722n , on the other hand if i install bullet 2hp on the same location it is working great tx ccq is about 50-70%. is that a devcei problem or a software issue.

messages are disconnected due to extensive data loss in logs.....

user is only 1 meter from the devices and alot of data loss and mind that my bullet 2 hp form ubnt is working great on the same data and mode.

user is only 1 meter from the devices and alot of data loss and mind that my bullet 2 hp form ubnt is working great on the same data and mode.

Metal 2 has a very powerful radio transmitter (1.6 WATT). If you are testing it indoors then I would try lowering the power settings and set the distance to indoors under the advanced tab.

http://wiki.mikrotik.com/wiki/Manual:Wi ... he_best.3F

Yes, that is correct. You must lower the power, or move away, until your Registration Table in RouterOS shows signal in the range of -55 to -65. For example -30 is much too strong.

Thanks to ALL for replys , ihave purchasedit for high power plus the problems comes when it is installed on the tower , where i have a another user antenna connected to my TPlink tl wn 651g card . ping is ok there but as i start to download anything ping goes to 3000+ms and download speed goes to 0 to 6 k over a 10mbit profile .

I undestand that you purchased it for this feature, but - too much power (and too high signal) will result in worse performance. Could you please show printout from the registration table when you have the issue?

Log just contains




disconnected due to extensive data loss

and no other messages.

this problem comes to other devices know ping goes high i am using mikrotik 5.19 for load balancing using pcc + hotspot+pppoe on same pc, users are connected via Local named interface to gigabit switch but there are no problems on wired network , problem just comes on Metal .

I have assigned tow ip's to same lan card like 10.0.0.1/24 for pppoe and 169.254.241.1/24 for hotspot for Local interface but my issue just remains to the metal and some times it comes to other wifi devices.............

CCQ is between 29-50%

distance goes to dynamic
72km

clients are just two

when the problem comes

i have claimed the warranty they are sending me new device if the problem comes again i will let you know thanks for replies plus


the older device gave problem just after opening the box and installing it on the tower.

YES - there is for sure a problem with the 2.4 metal !!!

I started with a mikrotik 433 and mikrotik 2.4 card (no problems) --- later changed to another brand (no problems) --- Then two weeks ago changed to the Mikrotik 2.4 ghz metal and that is when the problems began....

I have been everywhere in all the settings and the mikrotik 2.4 metal just does not work correctly!

I have 8 connections (802.11b only) (some are neighbors and others are about 7 miles away)
The distance reported by the metal mikrotik shows 48 and 50 km on some client connections. However the client mikrotik (RB411) reports distance of 7km.

There is something wrong with the AP mode of the Mikrotik metal 2.4 device (ver 5.22)

I ran into similar issues with Groove A-2Hn.

4 different antennas. Was able to get it finally going 3 days ago. ~45dB SNR and proper distance of 1km from tower was reported. Now 3 days passed, distance changes from 30km to 55km randomly. SNR varies from 25 to 35 dB with occasional drop in connectivity. (v5.14)

my post disappeared, so here is the recap

same experience with Groove A-2Hn (v5.14)

tried 4 different antennas, got it going finally, reported proper distance from tower (1km), SNR ~45dB, smooth performance

3 days after, distance fluctuate from 30km-50km, SNR from 25-35dB and there is occasional drop in connectivity

Ack setting in both dynamic mode (goes to ridiculous values) and manual ack (doesnt work at all)

I wish I could set manual ack but traffic fails to pass at all. (All versions from 3.x to 6.x) affected

access tp link & linksys messages are disconnected due to extensive data loss in logs....

i'm from egypt and i have same problem
please reply

Can at least one of you show the registration table printout please?

I have the same problem when I install the metal rb outside, but inside it works great even when signal is about 100db.

if the clients are not so far away then you can switch the distance setting from dynamic to indoors.
What is the ambient temperature where that Metal is located?
Have you tried to lower the tx-power a little bit?

I tried indoors instead of dynamic and lower tx power but it didn't work.
2 clients are about 100m and 2 about 500m.
The temperature is between 10 and 20 degrees celcius.

More than all of that, when I detach the meltal from the antenna away for about 5mm only , it works!!!
I tried longer pigtails but didn't do anything.

i know a little bit about metal but i noticed that you have placed 0 value for antenna gain,
changed to 28 and try,

regards

I tried to change antenna gain but nothing, the same problem still exists.
Also , I noticed that when connected as station to another mikrotik, the signal is unstable at all
all the time getting up and down from -51 to -83 and the ping requests timeout or high delay.

i have same problem with omni hyperlink..

so there is no solution to this problem?, I also experienced the same problem. I have followed the instructions to the lower tx power but still same problem.

If you have issues with wireless, start by resetting wireless to default with this command:

ros code

[admin@MikroTik] /interface wireless> reset-configuration

If this doesn't help, I suggest you to email support with your supout.rif file and we will try to solve the issue.

every thing done but same prablem

me too have this problem i have 7 device of Metal 2SHPn and didn't know why this problems happens
working perfect indoors but outdoors not working the client access is disconnected and some time connect again after that disconnected again try every thing the same problems
i asked every one have this device about this problem and they tell me that they have the same problem in there devices ...

we ask for the Solution , and how solve this problem

Thanks

I posted in my own thread , and now have seen this one here is a copy/paste of that thread.

We have a site that has been serviced by a 433ah with an XR2 for years. This had a 30ft run of LMR400 up to the 11dbi omni. It services about 60 Clients, and has worked great.



We saw some interference or what could be water in the coax. We have been elminating as much coax as possible over the past 2 years. So we took this opportunity to replace this with a Metal 2.4Ghz Groove.

However, this did not resolve the problem. Signals went from a range -50 t0 -80 and increased to a -15 to - 55. This is kinda crazy. I am not sure why there would have been such a huge jump in the -Rx signals of the groove. I am afraid something else is causing these weird signals or the groove is not reporting the sigs right.

This is a very old ap location, and we used to use Demarc brand cpe's. So about 85% of the CPEs are demarc with the other 15% mikrotik. I used the access list to only allow the mikrotik cpes to connect. Problem still existed of ping times from the ap to the cpes 700ms+. Even with no traffic. I switched from 802.11 to nv2, and then had PERFECT times, 1-12ms consistently with good traffic running. Once its switched back to 802.11b/g or b, the problem returns.

We have had cpes go bunk in the past and would go into compresion and spew out huge amounts of garbage RF. So this is kinda where I am leaning.

Now, when all of the clients are allowed and 802.11b/g is active, the problem still exists. At random times, the problem completely disappears in just 1-2secs, and ping times go <10ms for all the clients, and the interface starts passing around 3Mbps. This will last for less than a minute, and then the problem returns.

I have my methods of identifying the solution, but I would love to hear what others would do. It's been a long, cold, and windy day on towers and am burnt out. But i can't relax with this problem out there.

This is versions 5.21 - 6.0rc11.

Please give me some feedback, Thanks!

Adam

Must be Friday....hahahaol

you have the same Problem too Adam

Yup.......and customers are getting very upset. Will have to swap it out for something else.

Yes, that is correct. You must lower the power, or move away, until your Registration Table in RouterOS shows signal in the range of -55 to -65. For example -30 is much too strong.

Normis - The issue for me is that the signals where no where near that high with a different radio in its place. Also, NV2 Solves this problem. So what would nv2 be doing that would lend an idea to what is going wrong here. If I put the RB52hn or the XR2 back in place, the problem will vanish. Please advise!

Any updates out there? Did anyone get a fix for this? I am in dire need of a fix!

Downgraded 6.0rc11 to 5.1 remotely, and grove never came back......

I am also seeing 15dB jumps in receive signal, and I have lost all contact with it (ssh, winbox, mac-tel) etc etc.

Mikrotik? Can you please advise, I have submitted a support ticket with no response. I find it so odd how Mikrotik has avoided extending their support services to the same level that other vendors have (well just about all companies have). We have real businesses relying on this stuff. Please advise.

Mikrotik? Can you please advise, I have submitted a support ticket with no response. I find it so odd how Mikrotik has avoided extending their support services to the same level that other vendors have (well just about all companies have). We have real businesses relying on this stuff. Please advise.

Support answers within 3 business days. Tell me your ticket number here, and I will check the status.

sorry Adam for what happen to your grove

try this : http://wiki.mikrotik.com/wiki/Manual:Netinstall

and this : http://wiki.mikrotik.com/wiki/Flashfig

Hello,

are there any updates on this? We have exactly the same problem with an Metal shpn access point. Clients get disconnected while "extensive data loss" is logged on the metal.

I tried lowering the tx-power from 16 to 10 and then up to 13, which resultet in somewhat better dbi values for the connected clients but otherwise the problem persists.

All our RB33 boards work perfectly though.

Please Help!

cheers,

Tycho

I am starting to think that Mikrotik is blind to the concept that there is a serious problem with a Mikrotik product.

To to point - it does not work!!!!
Unless you call constant disconnects and high error rates and no throughput and distances 10x what a client really is, then it works just great (in the round file that is).

There is a serious problem, the problem is not software - the problem is the hardware - it is a bad product.

I am starting to think that Mikrotik is blind to the concept that there is a serious problem with a Mikrotik product.

To to point - it does not work!!!!
Unless you call constant disconnects and high error rates and no throughput and distances 10x what a client really is, then it works just great (in the round file that is).

There is a serious problem, the problem is not software - the problem is the hardware - it is a bad product.

Tom, did you email support? Please tell me your ticket number, and I will check the status. The forum is not an official ticketing system, it is a user forum, like it says on the top.

I am starting to think that Mikrotik is blind to the concept that there is a serious problem with a Mikrotik product.

To to point - it does not work!!!!
Unless you call constant disconnects and high error rates and no throughput and distances 10x what a client really is, then it works just great (in the round file that is).

There is a serious problem, the problem is not software - the problem is the hardware - it is a bad product.

BAD IDEA. ..

Everybody with Metal 2SHPn problems, please write the following:

- did the problems start right out of the box, or it worked for some time at first?
- did you properly connect an antenna before you powered on the unit?
- if yes, what antenna did you use exactly ?

Everybody with Metal 2SHPn problems, please write the following:

- did the problems start right out of the box, or it worked for some time at first?
- did you properly connect an antenna before you powered on the unit?
- if yes, what antenna did you use exactly ?

- yes, the problems start right out of the box, it worked sometime but ping are not stable, sometimes above 2000ms, and get timeout. many client are disconnected with info= extensive data loss...
- yes.. i'm using omni tplink 15dbi... attached in 2 pipe (+-11meter height) with wds client tplink 5201g which only +-600meter distance from tower

have try to lower the power and try many setting... but the problem still occured.

and there's an experience, theres laptop using wifi atheros chipset(bgn) which cannot connect, it just about 10mtr from tower, but then i set the metal to 20mhz channelwidth, and then it can connect. i dont know whats the difference, while the other laptop(mostly using broadcom chipset) are just fine when the metal is set into 20 40mhz

hope theres solution for this... have been a month finding the answer.. thanks

i asking for solution from long time and nothing happen distance between the Metal AP and the wireless clients less than 1 km , i'm using 12db omni antenna in Metal and trying another antenna Hyperlink and the same problem , update firmware in wireless clients the same problem , try many setting ,
u must know many peoples have problem with Metal 2SHPn

A problem called 'ver. 6.xx' coz it's buggy.

A problem called 'ver. 6.xx' coz it's buggy.

i'm back to v 5.24 and nothing new the same problem clients access disconnected or high ping 3533ms and request time out

Everybody with Metal 2SHPn problems, please write the following:

- did the problems start right out of the box, or it worked for some time at first?
- did you properly connect an antenna before you powered on the unit?
- if yes, what antenna did you use exactly ?

1: Yes, right out of the box
2: Yes
3: connected to 12db grid before turning on

seeing spikes from -65 then back to -85, erratic ping times

Does any one of you have more than one Metal 2SHPn? Do all of your metals have this issue, or only some? Can you send the defective units directly to MikroTik for inspection? Please contact support@mikrotik.com to arrange this

ba7abak, are you receiving our emails? I see that we have six closed support tickets with your email. Did you try what MikroTik support suggested?

Yes normis trying what MikroTik support suggested to me and nothing happen the same problem

you ask about every one have Metal 2SHPn have this problem i think Yes Yes , i'm from Egypt we talk in Arabic forum and try to repair any thing , change in setting , change Antenna , the power adapter with high volt-amps and nothing new the same problem

Does any one of you have more than one Metal 2SHPn? Do all of your metals have this issue, or only some? Can you send the defective units directly to MikroTik for inspection? Please contact support@mikrotik.com to arrange this

I have two of them, the other one is working perfectly. The other one is running in ap mode with mixed clients, and been solid.

i update to last Mikrotik v6.0rc12 , nothing repaired ((((((

Just when i was thinking of buying a Metal 2 for my home AP. I will reconsider my decision now and wait for a while if they fix it.

i don't know why they didn't fix it many people have problem with metal and nothing repaired :S

i think the problem Software not hardware

injured x i want your help to connect tplink 5210 with wds mode
i am using metal 2shpn with tplink 5210
problem is that when i power on metal it is not connecting with wds automatically
everytime i have to manually disable then enable wlan interface to connect with wds

i will pay for this work

injured x i want your help to connect tplink 5210 with wds mode
i am using metal 2shpn with tplink 5210
problem is that when i power on metal it is not connecting with wds automatically
everytime i have to manually disable then enable wlan interface to connect with wds

i will pay for this work

i need to know your network configuration, or pm me.

injuredx or any friend here can help me too

any one add my email : ledonet@ymail.com

to make Metal configuration by team viewer

wds problem : dynamic bridge port

this shows one of the wds ap are disconnected... tx/rx signal gone... need to be restarted again to connect...
i'm curious why it always happen, then i set and remove that ap manually from wireless tables - registration
and this is the log...

Normis, sorry I didnt see your posts on here. To your questions, I have 2 of them, both with the same issue, both had the issue out of the box.

NV2 resolves the problem. Users, if your clients are all mikrotik, nv2 will resolve the problem. I am not that fortunate.

I can ship you a defective unit, I also still have a defective unit in the air with a 10.5db omni. Its mounted directly next to a 411 with a rb52hn with a 13db omni, when the clients are moved from the metal to the 411, things work fine. I have replaced the omni more than once.

The suggestions on my ticket are just busy work for me, they are not related to any fix. In addition, the response on tickets are more than 2 weeks in between responses.

I really wish mikrotik would invest in some sort of tech-support, even paid support. However, the only time we need it is when we have to prove to Mikrotik that they have a bug/issue.

theWISP, please contact support on the ticket that you have opened so we could follow up on your problem.

any solution?................

No solution, the 'ticket' is a joke. its just being used to keep the problem off of the forum. They keep asking question ive already answered and giving me nothing to try. they asked if i had a unit to send them, i replied yes multiple times but with no response.

I believe it can be safely assumed there exists a problem that can not be solved with a software change, and a recall of hardware is not an option.

As my companies needs expand, I have come to the conclusion Mikrotik can not be considered carrier class without the proper support infrastructure.

Adam R Johnson
President
theWISP.net
1-815-304-6008 x211

i also think that Mikrotik need to open an online ticketing portal
with reply time max 12 hours

market is getting competitive and with this kind of support it's not going to be carrier class

i have placed 433 with R52hn and hyperlink 15dbi omni at 70 feet mast(village area)
i was not even able to detect the signal from 1km distance(25feet at client end) with TP-Link 5210G
noise floor is: -119

Hi all.
I have a similar problem as the others.

I have set up a mikrotik metal 2shpn as AP. I connected a TP-link 12Dbi OmniDirectional Antenna on top of it
The configurations for power are set as default. The Band set is 2GHz N with channels 20/40Mhz HT Above (wich I don't really understand).

For the first day a client about 150 m away was connected and working flawleesly with signal -68dbi to -72dbi and CCQ average 88-92%

The next day he can't connect at all with the AP. Looking at the logs I see something like this:
00:18:06 wireless,info F0:7D:68:65:17:7B@wlan1: connected
00:18:22 wireless,info F0:7D:68:65:17:7B@wlan1: disconnected, received deauth: class 2 frame received (6)
00:20:16 wireless,info F0:7D:68:65:17:7B@wlan1: connected
00:21:05 wireless,info F0:7D:68:65:17:7B@wlan1: reassociating
00:21:05 wireless,info F0:7D:68:65:17:7B@wlan1: disconnected, ok
00:21:05 wireless,info F0:7D:68:65:17:7B@wlan1: connected
00:21:10 wireless,info F0:7D:68:65:17:7B@wlan1: disconnected, received deauth: class 2 frame received (6)
00:48:08 wireless,info F0:7D:68:65:17:7B@wlan1: connected
00:48:32 wireless,info F0:7D:68:65:17:7B@wlan1: disconnected, received deauth: class 2 frame received (6)
00:48:38 system,info device changed by admin
00:50:25 system,info device changed by admin
00:52:20 system,info device changed by admin
00:56:50 wireless,info F0:7D:68:65:17:7B@wlan1: connected

What do you think the problem is?
For what I've searched i see that i must decrease power of the metal because is very high.
I don't really understand because when the wireless device was connected the whole day, it's signal was somewhere at -72, -75dbi, and the TX power was set as defualt 32 dbm. As long as i know this isn't to much signal


I tried lowering the Tx power to 26Dbm and now i see that the client is connected and the signal "jumps" from -40dbi to -61 ,-65dbi to -81dbi with laptop.The overall tx ccq is somwhere 16-21 % with distance 1km.
Whith smartphone client can't connect. In the logs is shown smth like this:
01:13:23 wireless,info CC:FE:3C:CC:81:E0@wlan1: connected
01:13:50 wireless,info CC:FE:3C:CC:81:E0@wlan1: disconnected, extensive data loss
01:16:56 wireless,info 40:D3:2D:AA:74:03@wlan1: connected
01:17:36 wireless,info 40:D3:2D:AA:74:03@wlan1: disconnected, extensive data loss
01:25:33 system,info,account user admin logged in from 192.168.1.86 via telnet
01:25:48 system,info,account user admin logged out from 192.168.1.86 via telnet
01:31:11 wireless,info CC:FE:3C:CC:81:E0@wlan1: connected
01:31:42 wireless,info CC:FE:3C:CC:81:E0@wlan1: disconnected, extensive data loss

I have to say that this is the first time that i must lower tx power to get better results, it's kind of weird

ba7abak, are you receiving our emails? I see that we have six closed support tickets with your email. Did you try what MikroTik support suggested?

I've downloaded package update metal but after he was taken to the metal prepared run no longer works, and I can not access to it please help
4 Devices gotta Metal stopped on the way

try doing a netinstall

Everybody with Metal 2SHPn problems, please write the following:

- did the problems start right out of the box, or it worked for some time at first?
- did you properly connect an antenna before you powered on the unit?
- if yes, what antenna did you use exactly ?

Yes, straight out of the box.
Yes
12dBi TP-Link Omni

I have same problem...
In building work perfect, but outdoors lost connections and I noticed strange Noise Floor is -97, but one time on 2-3 sec change to -40-45db, and after that came back to -97.
(new from box, 3 different antenna's)

Add video
http://www.youtube.com/watch?v=uw9IjKKE ... e=youtu.be

Hi there

We have had 2 Metal groove with the same symptoms.

Tried another stock Metal-2SHPN exactly the same, straight out of the box.

We bought 7 and all of them are not working properly. The signal goes low and traffic go to 0.

Any idea?

It look's like support is on vacation...

No word from MikroTik about it ?

I am having these same problems out of the box. I have made several tests and I have found the following...
When connecting to using an N mode client usually everything is ok enless The client cant talk back well...
When connecting using b or g mode client it jams up the data on most all connections.
So I started reading and adjusting setting compairing with another branded older box from 3 miles away using b g since this box does not do N.
The settings I am testing are listed below. please try and let me know if it helps or impoves yours as well. the options with * are settings I changed to try to resolve speed issues
mode ap bridge
band bgn
channel width 20mhz
wireless protocol 802.11
freq mode manual
no country set
ant gain 0
dfs mode none
post 2.9.25
*wmm enabled
bridge mode enabled
*multicast helper full
rate selection advanced default
max station 2007
*distance manually set to 5 << set to the correct distance of furthest client
*periodic cal enabled
*cal interval 1:00
*hw retries 7
*hw frag 2346
*hw protection mode rts cts
*hw protection threshold 2346
*frame life 0
*adaptive noise immunity ap and client mode
*preamble both
*discontect timeout 15
*on fail retry time .30
wds mode static
wds bridge I have set to bridge1 since I am sever to other repeaters
wds cost 100
wds range 50-100
wds ignore ssid is checked.
Transmit power mode card rates
tx power 32 this is set for testing only please set this to your legal limit or need power.

*Update*
On other branded unit previous settings helped but i still had data loss and some ping timeouts.
Could do good speed test result but streamming video for several minutes kills data thu put and could not
Get connection sold. Today i switched this other brand for metal 2 and problems are fixed.
Other settings that have been changed are...
Noise thrushhold -105
Add new item to access list
Mac blank
-120..-80
No check auth
No check forwar.
This setting does not allow clients to connect with week signal and should kill clients that go too weak to talk.
Other problem.. wds low data at 2 miles but client mode works great. Wds wasnt needed anyways.
I tried changing noise thrushold to -129 to test and t can hear very well but causes to much noise in reciever.
Could not get good data on -129 setting
I will do some more testing with other branded unit here local since i have the unit and i will post my findings.
Hope this helps others. I think issue maybe wireless driver related causing issue with some other chipsets.

Today i did alot of test while streaming video at 5mb. With power setting at 32 data goes out after 30 minutes or so
I changed setting to 28 and seems to be rock solid for now no disconnects and stream is steady after 60 minutes.
I also found that bgn mode causes hangs and low data when mixed. So i tried n only mode. Same issue. Tried bg only fixed the problem. I have now switched all my boxes to this and then edit access mode to -120..-90 for now to test this setting long period of time with mixed signal levels from differant clients.

Today ran into issues again, went through 2 linksys routers differant models. was having internet hangs.
sometime looses internet all together. I decided to manually configure the cable modem and remove the router since the cable modem has built in router. Only problem is the internet company may change my lan ip. they have done before. I believe some of these problems maybe in some settings in the motorola cable modem.
model# SBG6580
I logged in and changed wireless mode from 2.4 to 5ghz to get it off of the 2.4 band.
Filter Proxy disabled
Filter Cookies disabled
Filter Java Applets disabled
Filter ActiveX disabled
Filter Popup Windows disabled
Block Fragmented IP Packets disabled
Port Scan Detection << was enabled and I disabled it.
IP Flood Detection << make sure this is disabled
Firewall Protection << was enabled and I disabled it.
I read some posts about other that had issues and this seemed to fix their issues.
Also I need a script or need to know how to disconnect clients with week signal
because this seems to be a big issue with why my access metals are having issues.
Today i found someone connected that had good signal but then walked around and signal went low or dead but
was still connected causing my ap bridge to freeze with no data. I found a couple scripts but I do not want to
input something bad as I am trying my best not to climb to the top of this tall tower! thanks

rock solid 12 hours
A setting i changed in metal i forgot to list when trouble shooting from my last post was hide ssid unchecked.
Now no clients see to connect with low signal fixed data drop outs and no lost packets. Need that low signal kick script.
Thanks

the problem solved about the Metal 2HSPn or still exist? Kindly let me know because i was going to purchase 100 Units but through the forum i found with it the unusual issues. If mikrotik fixed that issue then please let me know thanks

Rock solid connection at 2.6 miles using two 24db dish from cali amp. These dishes are 7 years old working like new still.
I have had ping -t running for almost 24 hours and not missed a packet. In fact I have watch dog turned on waiting for a missed packet and it has not restarted. ping time is usually 1ms and sometimes 2ms. every once in awhile i see 11ms to 29ms but this is nothing to worry about as it maybe switching speeds. I am going on vacation on friday for a week so I need it solid or to where it can restart to fix itself while i am gone. So I may not do anymore testing for speed increase untill i return.
As far as i can tell, I am very happy with unit even though lack of people helping in forums. Most of the info I have learned was trial and error. Keep i mind there is access to so many settings in routeros that it can be easy to mess it up. I would sugest not to use opera web browser for web login. and once set to correct mode ONLY use WINBOX!! I had a box brick using opera browser over the wireless. then I had to netinstall to recover unit but was success thanks to baltic tech help.
I am still waiting on what to set to kick clients when signal goes lower then what I need because it causes unit to hang data.
oh and speed test today at noon using web site on a laptop connected to another wireless router inside clients house 15mb download and maxed my upload at steady 5mb upload from client. the speed may have been more but on lunch break didnt have time to play around since i forgot to change the ip from manual at customer house. hope this helps.

Here is a screen shot of ap bridge running with some steady data moving. maybe video streaming. The ping window is of the client. Connection time is correct, I rebooted today at around 12pm, before that it was 10hrs or so no loss.

Back from vacation... current settings worked while away and was solid. I yesterday I connected connection #2 and had both connections streaming video, I then noticed very high ping and low data, including connects and disconnects. So I did a lot of reading and found the issue. Legacy B 1,2,5,11 using mixed mode was causing a lot of problems. This is not a mikrotik, router os, driver, or chipset issue... but is in fact a known issue with all wireless connections. Basically N knows when N is talking and tells others to wait till ready. G also knows when N is talking and N knows when G is talking. But when B is talking on a N and G connections are alive it cant hear both because B clients do not know when other N and G clients are talking because they can not hear them and B does not tell them N or G is busy. So if your mixing bgn then it will cause issues if all are using data. Here is the testing settings for this. Keep unit in BGN mode. In advanced mode, Data Rates tab,
rate selection: advanced
rate: configured
supported rates B, none checked.
supported rates a/g, all checked.
basic rates B, none checked.
basic rates a/g, only 6mb checked.
the next setting I noticed speed increase. HT tab
ht tx chains: checked
ht rx chains: checked
ht amsdu limit: 8192
ht amsdu threshold: 8192
ht guard: any
ht ampdu priorities: ALL CHECKED
Now all these settings are applied to my unit, I can connect with client with others connected and I see very little change in pings to others connected. No more data loss during new client connection to already connected clients.
Hope this helps. I will post with any new updates.

Tonight I changed the distance setting from my manually set number to dynamic and it now reports a correct km number!
This increased some speed and slightly better ping!

using a total of 130 pieces of metal 2. I prefer the sector as the antennas. the same way as omni, grid, .... memlunum I have no problem with a lot of metal from 5 to 2, and metal. I would always check CCQ during installation and always try to avoid places that will create interference.

Today I made some testing with the advance settings...
I changed all advanced settings to mikrotik default values and everything went well until put under pressure
of a speedtest for long period of time. After several minutes I lost data to almost nothing and had disconnects
and reconnects. I changed everything back to my settings and everything was back to normal. I would suggest
for the default mikrotik settings in firmware to be changed to help new customers make connections with less
issues.

I have no problems for the last 2-3 days, with the v6.2 firmware upgrade .
Everything seems to work properly tiil now with the default configuration.
Still waiting to see if any other issues will occur with more connected clients the next days.

I don't know if this will help anyone or not:

I was having similar problems with my Metal 2Ghz model with the factory installed v5 of the ros. Same issues with an upgrade to 6.1 as well as 6.2.

This was very frustrating until I discovered someone on the site using some older off brand 2GHz outdoor wireless cameras about 300m from the device. With the devices turned off the connections became stable and speed increased to what was expected for the available bandwidth.

I was able to diagnose the interference using the Dude software on a laptop which was plugged directly into the PoE.

It seems to me that the cameras were making a non-stop signal "drowning" out my other Metals with Omni's running WDS network.

I also increased the disconnect timeout to 10s from the default 5s.

gkrueger, was your end test using 6.2 firmware?
I am still on 5.25 firmware. Currently I have changed from BGN to G only setting. B only does not work or has very bad data loss. N Only does not put out faster data. G only connects fast and data is good at 2+ miles. This morning I had a metal radio lock up. I went to the log and it says client connected then had data loss and disconnected. after that my 2+mile metal disconnected, and then everything else froze on the wireless side. Nothing I do seems to fix that problem.
Although, I have noticed that the G only mode works a lot better and faster then my old branded unit.
I need to know if upgrading to version 6 firmware will help with issues?

Scott,

Actually I have currently:

2 x Metal 2GHz purchased a few months ago that I upgraded from 5.25 *Software* version, *firmware* is a different version. I have not altered the *firmware* on any devices. Upgraded to 6.1 version of the OS via netinstall.

2 x Metal 2GHz purchased a few weeks ago upgraded to *software* version 6.2. This is when I noticed that the *firmware* version had a newer release version so I have been thinking about upgrading the *firmware*

I'm not clear on the difference but I do know the firmware is different from the software version.

So to recap:

2 x Metal 2GHz with 9dBi Omni antenna
Software v6.1

2 x Metal 2GHz with 9dBi Omni antenna
Software v6.2

I am still working with configuration of the 4 devices but I have the WDS Mesh working on a single subnet. I have noticed that the Portal to the Mesh that is the only device connected via cat-5e to the main router which is connected to the "internet". Anyways, this device will stop "working" with the wireless like it does not show on the list (SSID) but still responds to winbox (shows in the winbox ... list after a delay). This has always been fixed by a power cycle, which it comes right back. I have also noticed when the wireless doesn't seem to be working (showing on the SSID list on client) it's actually busy using the wlan device to scan or snoop or some other tool or once I had set it to "wds slave" mode for the wireless mode and it was scanning for available WDS devices so it wasn't showing on the SSID list.

This is my first time using RouterOS devices and I am a somewhat experienced wireless (wi-fi) installer. I have just used off the shelf Belkin or Linksys devices up till now. I was rather worried about my first experiences with these Metal devices but now it seems maybe all my issues were caused by either interference of outside RF at the site, or bad configuration.

I will upgrade to 6.2 within this next week and test it. I will report back with B only, N only, G only, and MIX BGN.
gkrueger, don't forget about the watch dog. The watch dog will ping an ip address on the wireless and when it stops pinging it will reboot the device for you and record the log if needed. This is saving me from unplugging. I tried 4 hour in the setting and it was to long. so I have mine set for 1 hour. The only down fall of the reboot is the metal stops all my other wifi devices on other channels while it reboots, and this can cause the other devices to reboot as well since it causes no ping for the watch dog. But at least they are close to being is sync with the hour.

I have 2 x 2,4Ghz Metals in a 9kM link on 25dBi mesh antenna. With default power setting signals into -30's on the other end!
Set radios now to -23dBm to get signals in -50 to - 60dBm range.
Station-bridge on one end, bridge mode on the other. (So a bridged link running over them.)
Both running v6.2 and updated firmware.

While in default power on 20/40Mhz NV2 channel 14 (Japan) units produce a stable 150Mbps conn.rate with an over 100Mb aggregated tcp traffic, units collapse each time after some 20 - 50 seconds.
Its usually the client that needs a reboot to get it back working again.

I have been playin now for over a week to get the link stable but without success...
Tried to swap AP-station, same results but now the other units needs the reboots...
Tried to use nstreme but don't seem to get any reasonable traffic over the link.
Tried with fixed conn. rate and with default; same results.
Tried to lower power step by step. Problem persist with result that powers lower than 21dBm units wont talk to eachother no more while the are still connected with -65dBm and 150dBm conn. rate!
Tried units by day, but also by night and in the rain. Same results so temp. can't be an issue.

When idle the links stays running for days. But everybit more traffic than say 5Mb above, station drops and need a reboot. (just disabling->enabling wireless is not good enough, soft reboot is needed.) Sometimes also the AP wireless crashes... (radio stops because I can't even do a scan anymore to 'see' what can be seen. Scan stays empty.)

After reading all these post here and hearing some other similar stories from a friend of mind I am thinking MT should actually stop selling these units or solve the problem.
The problem itself is imho the radio's. The are not stable and under significant load just collapse.

I think these units have to go RMA....

OH, by the way. The link runs over a town with therefore several hundreds of wifi signals to be picked up. Some with strengths upto -50! Hence I use channel 14. I am the only one there.....
But since the channel 40Mhz wide I am basically covering the whole 2,4Ghz band.
So yes, interference can be the issue. But if the metal is not able to handle this it is not even worth putting it on the market! For empty 2,4Ghz spectrum you have to fly to the moon! (And probably still picking up signals from us....)

Rudy, start with supout.rif file made at the moment when "it's all bad" and we will see what is happening. It could be software, could be hardware, but email that file to support first.

I can only make supout.rif after the reboot, when its all working again... it that any good?
I even set watchdog to produce one if the ping to the other end of the link fails.. but watchdog doesn't produce any files. Basically it means the unit really crashes, isn't it?

If the supout.rif from after the reboot is ok I will send that to support?

we need the support output file which is made before reboot - when the problem is happening.

we need the support output file which is made before reboot - when the problem is happening.

Ok, so I make the support output while I set the unit such it will crash.. it usually happens within 30 secs... not enough time to make a support output. But I'll give it a try....

This is one very interesting topic. Not because Metal 2 obviously has issues but because support@mikrotik.com is not working/replying at all.
I have very similar issues with a few Metal 2 devices I have bought that are loosing client registrations until rebooted and managed to capture the supout files before reboot while in a frozen state. Sent them to the email address above and haven't received an answer at all for 10 full days now. Is there maybe some other support email that IS working ?

As for Metal 2SHPn, I have found out that firmware upgrade from stock 3.02 to latest 3.09 helps in a way that problems don't occur for a few days, maybe even a week or two, but eventually it will happen, so it doesn't solve the problem. For me, the workaround is to schedule a nightly reboot of the device ...

Thanks,
Ivan

does the problem with Metal2SHPn‏ still exist ?

Well, mine's are send back via my dealer to MT upon MT's own request: "probably hardware failure".

So I'll wait until I got something back and try again...

it works great even when signal is about 100db.

Hi scott, any news with update 6.2. I just got 5units metals 2.4ghz and im about to upgrade before deployement

Sent from my GT-I9505 using Tapatalk

it works great even when signal is about 100db.

What is 100db?

Anyway, they work great without traffic. The more traffic you put on, the sooner they loose the link.
20Mb of traffic runs for about half an hour and than the link collaps.
50Mb takes about 5 mins to make it collapse.
Max. speed was about 100Mb aggregated (So 60/40, or 70/30 etc.) up to 125Mb aggregated takes sometimes less than a minute to make the link drop.
Sometimes a disable/enable was enough the bring the radio back online, sometimes a complete reboot was needed, and on some occasions that was not even enough, it needed a power cycle reboot.

I've done several tests with them and showed all results to MT and the last word of them was: Send them RMA because probably hardware failure...

So, the units are shipped back through my sales channel (so it will take some time before they reach MT) and I hope to get some better performing units back. I will tests these under exactly the same circumstances so I'll find out soon enough if it is really a hardware failure (which I doubt) or another kind of issue.

My suspicions go towards the chipset/software combination. The link I use is NV2 with encryption on a very clean frequency (out of band) and when it drops it is each time because of NV2 errors.
When the link is made in a crowded spectrum (so in the normal 2,4Ghz band) it even lasts shorter.

Maybe also the software. I got very strange reading of the signal at time. Although the link signals were normally in the -35 to -40 range (7km link!) the sometimes showed -15 or even -5!
And in reducing the power level I noticed a change in outpot of 20dBm when changing between 23 and 24dBm power output...
So 1dBm change in power output made the other end signal reading jump with 20 or more dBm's... not good.....

problem all metals - pure cooling system. there absolutly no radiators on the atheros chips.


They are extremaly hot (ithink more than 100*С) when bandwith goes to 20-40 mbits\s.

Thermal sensor in metal is stupid. even outdoor is 10*C , metal just power on - he shows 33-35*C wtf?
here 2 pics. 1 pic - metal, 2pic - bullet. did you see the difference? main chip on the bullet is with radiator.

mtik have 2 atheros chips and both without any cooling system.

WEll, I agree on the temp issue. Already noticed that both during a hot summer day in the sun (metal felt really warm-/hot in the hand) and running or in the middle of the cool night the temp always is around 35 degree celsius.....

And yes, heating of the chip would explain why this so cold "industry standard" device crashes when it really has to perform (= traffic load)

My units are back for RMA on request of MT. "Hardware issue" was their argument. Lets see what they come back with.....

cooling system ... is there any space in metal to do something in the cooling system.
i have ordered 2 x metal, when they arrive let's see what can be done .

I have many Metal 2Shpn installed on different spots and the problem appears on some of them. Do you think that some heat conductive gel/foam placed between the case and the chipset would solve the problem?

I've found this from Fisher Electronik:
http://www.fischerelektronik.de/web_fis ... ndex.xhtml

It's a 5mm heat conducting compressible gel designed to be placed between electronics and heatsink. Considering the Metal's case as the heatsink i think it could resolve the problem. There's also from the same manufacturer a 6.35mm foam but it seems it has a lower thermal conductivity.

3M has something similar. If I can find where to get some I'll post the results.

The other side of the board is connected to the Metallic enclosure, which works as a giant heatsink. CPU doesn't need a heatsink, since it doesn't get very hot, and since the case works as a heatsink from the other side of the PCB. The issues you are seeing are probably related to something else.

Then you should tell us where's the problem because I've got an stock of about 20 Metal that are stuck there waiting for a solution. I've installed about 60 of them this summer and I've found the error in about 20%. We are replacing them for Grooves as soon as they stop working. Sometimes the problem can be minimized with an scheduled reboot every few hours but is just a temporal fix. We are not installing them by now waiting for a proper solution. If at least they failed from the begining we could RMA them before installing but the problem doesn't appear until they are on a production enviroment.

Then you should tell us where's the problem because I've got an stock of about 20 Metal that are stuck there waiting for a solution. I've installed about 60 of them this summer and I've found the error in about 20%. We are replacing them for Grooves as soon as they stop working. Sometimes the problem can be minimized with an scheduled reboot every few hours but is just a temporal fix. We are not installing them by now waiting for a proper solution. If at least they failed from the begining we could RMA them before installing but the problem doesn't appear until they are on a production enviroment.

Please contact support@mikrotik.com and include the Serial Numbers from the Metal2 boards that have failed so we could check.

hi
1st sorry for my english
me too using metal 7 peces
i have no problum with metal but except one
after 4 or 5 days running metal is going to hang its wireless is hang only not fully routerboard when it is hange u can scan for ap,s & there is nothing after reboot manualy its working perfect for 4 days only and then againg reboot it
our all stations connected via metal5 with grid 30db n there is no problum in metal5 but problum is only in metal 2Ghz
i have too send ticket to mikrotik butt i get no responce
i suggest that we should move on to some other brands
i have alsoo buy 6 peaces of rb SXT G2Hnd also 1.6W
it has too some problum butt all is scrape now no responce from mikrotik
i m thinking how can mikrotik close his eyes for his customers
repply pls if any one got any fix for my problum

Great!
I have the same problem. I was changed my metal 2 to a new one yesterday, but the problem is the same. After about few hours the clients are disconnected until i reboot the metal.

It's amazing, all metal and I will send you back to sue for damages.

1 weeks before I realized that was just standing against a MikroTik problem again, you do not have a solution.

DO SOMETHING!!!!!!!!!!!!!

2 weeks ago I DID THE MISTAKE TO BUY 4 Metals...

"Mikrotik Support", who is gonna give my money and wasted time back...
I decided to remove 4 Bullet M2 connected to 13db omni antenna and working as an AP for clients in out hotel and because I was* a fan of Mikrotik and I needed do create a second "virtual" wireless interface for my purpose....
I did what, I found Metal as a competitive model to Bullet M2, even with better resources.... so I paid the money.... 1 week of problems and I put the metals back in the box and returned the bullets to the antennas.....
The problem: When we have 5 clients connected at 6 a.m in the morning we dont have problem, after 7 a.m till 12 p.m (when we have more than 10 clients connected) we have pings going to 1000 ms and more, ccq goes down to 5-10% and INTERNET??? No Internet.
in the other hotel, we have a RB433AH bought 2008 with 2 Mikrotik cards and 2 sector antennas.... 2008!!! and still working very well

What do we have here, RB Metals bought 2 weeks ago working as a shit !!!
Now my company has wasted money and even more, we have make our clients not happy with is, because mikrotik is afraid to say
DO NOT BUY RB METAL, cause we have an issue!!!! Because we are a small company and we are not able to test out product before selling it.... Is this your problem???

Iam not happy with you...
this topic has been started 2 years ago.... and u have no idea what is going on.... and haven't mention officially about the problem....
I want my money back or your product changed for a working ONE... What should I do ?

Hi there

We have one site which has 3 Metal 2Hn, and 3 Metal 5HPn. All three the 2.4 has stopped working. We have replaced it with 433 boards and radios.

Hello Support.

I am an old hand radio guy. I have done radio for more than 50 years. When I first looked at the 2SHPn I thought it would have a cooling problem. I should be able to help you do some design modifications to help keep the internal devices cooler. Yes, I am sure this thing would melt in the sun in the Sahara. Outdoor enclosures are not simple. You need a sun shield and fins. A higher cost high temperature environment model might be a good idea.

I have had the radio shut down. I came here trying to find the data sheet on the radio chip you use. I have not found it yet. I am pretty sure the signal did not shut down last night due to heat, as it was cold here, and max traffic speed should not greatly exceed 8 mb/s out. I think there is another problem. Could it be SWR fault shut off coupled to a not very good antenna? That could be answered in the data sheet. In any event, when you get to 30 dBm you SHOULD have a directional coupler on the circuit card with a read out of reverse and forward AVERAGE RF power. That is another thing I might be able to help you with....

Still.... I think the problem may well be in your OS, as when this happens to me, I cannot generate an SUPOUT file. It jams. So.. that is a clue. I run it as a local HS with up to 5 uesrs. Last night it jamed up and off the air with no more than 3 users. It was cold out (approx 0 degree C). My next experiment is to move Hot Spot control into a different RB.

For the purposes of this forum.... the problem of reliability of this unit. continues.

Oh yes... I forgot to mention.. My OS is the most recent: 6.15.
This is a problem that goes back to the OS installed when purchased, a version 5.x.

Hi there,
Unfortunately, I am one more of the many affected, I have about 40 Metal installed in different locations, some work, some do not, and if they work, sometimes fail, not like the others, but once a month, for sure.
The problem specifically is that for no apparent reason, they are disconnected from other Metal (WDS) when these are linked.
Sometimes when the connection is cable, continue responding to ping, but not distributed, the WiFi signal to customers.
an
I've tried everything, Schedule reboot, every 3 hours, replacement of parts, such as power supply, or poe, RJ245 cable, antenna ... etc..
I have also purchased voltage regulators, on suspicion of peaks in the electrical network.
But nothing works; temperature is certainly an important factor, as this equipment, which are installed on the outside, but really do not have clear, that this, is the determining factor in the malfunction.
Right now I have an open Metal, and indeed does not have a heat sink on the (Atheros) CPU, will try to make one up and try.

If anyone finds a solution, share it, please.

Thanks

Sorry to say, you can only use them as CPE units. Replace with 433 and radios, the only real way of a solution.

I agree, something is not working correctly with this hardware.

http://forum.mikrotik.com/viewtopic.php?f=3&t=86804

Hummm... Power back off to maximum 30 dBm max did not cure this problem. Today, I had 4 outages. This is greatly more often than before. As of today I added to the set up a PPoE server... so the OS has even more to do. That may explain all the outages. Just before a reboot it did make a SPROUT file today, which it often has not, and maybe Tic support can look at it and find out what is happening. As of now, I suspect the OS is sending some wrong signal up to the chip. Only my guess.

I am thinking Mikrotik farmed out the design of the metal to a third party, and no one there knows enough about how it is supposed to work. I was told they do not have a data sheet on the radio. Without proper documentation one cannot design the unit.

Proof that there is a serious issue with the 2SHPn.

Hi. Mikrotik support has asked me to update the OS and have done so from the unit as shipped, and then for for 6.15 and .16 and .17....yesterday I installed new firmware. No,it did not help, it shut itself off again near midnight local time.

It is interesting that it receives but does not transmit when it has failed. 802.11x is a time division duplex system, meaning the transmitter must shut off and on thousands of times per second. I have scanned the spectrum with another 'Tik while the Metal is in failure mode, and when I scan I see there is no signal on the channel or nearby channels. This was done using a Groove sitting nearby inside this home. If I can trust Mikrotik on its spectrum analysis, it means there is no interference, no malformed signal, no nothing on or near the frequency. Yet while not working the radio can recieve the weak cell phone wifi signal from a neighbour. This leads me to suspect Qualcomm as the point of failure. It is consistent with a transmitter turning off as instructed but not turning back on when instructed. I just wrote support@mikrotik that I will no longer work for them trying to debug this.. for free. The metal is in use, outside on a short stick, and in use by myself and 3 neighbors, and I really want to link up someone who cannot get internet who is dying of cancer. So I have a real DEADline.

I want them to just send me a replacement unit at their expense.

By the way, Qualcomm will not send me a data sheet for the radio chip... forget it. They do not answer email to a mere user.

I just noticed they just posted a 6.18 os... maybe I will upload it and try as I can do this from here with the cat on my lap..... but really, if it is a chip failure then what good is a firmware or OS change? I would be more impressed if sent a software tool that can debug the radio chip.
***************
Status 24-July-2014 20h28 UTC+2: : Kristaps at support said, "Please, download and install the latest pre-release nightly build:"
I did, I rebooted, it was installed as when I logged it said it was a release candidate, etc.. and tried to register to the Metal wifi. After 3 attempts I knew this release candidate is faulty. I saw the beacon and the signal was strong. But I could not register. So back to 6.17 for now. Phew. I wrote an email to the neighbors with whom I share wi-fi... trying to explain the almost 1/2 hour off air. Enough.
***************
Just noticed upon reboot it did not install 6.17 but is running v6.18rc4 and working thus far.. at 22h42. Two reboots ?
***************
v6.18rc4 still running since ~ 9 pm yesterday with PPPoE and other crash causing services. Much better than before. I will run a stress test later today.

I am flagging this Maybe Mikrotik has fixed it in the RouterOS v6.18rc4 built yesterday at 13h58, on top of firmware 3.17. It has been running almost 22 hours, PPPoE and Hotspot with attached users with no problems. I ran a bandwidth at test to my Groove for about an hour @ 162/145 mb/s, with no problems. Before now, this sort of stress would make the unit quit after only a few hours if not minutes. If after a week if it does not quit, I will believe it is now reliable.

My main complaint is that their tech people did not communicate with us. It is as if they fear admiting any of their products can be imperfect.

It looks like they have fixed problem where the wireless goes to 50% and stops responding but there is another problem. I noticed that the wireless will stay up and the registration will show the 2nd device connected but you can't ping the 2nd device. You have to stop the wireless and start it again on the 1st device. Once you do that it will start working again for awhile but it will eventual break again. I can reproduce this by doing a bandwidth test and setting the direction to both. After some time it will break.

My Metal has 4 or more registered devices all last night including this Dell laptop. When the AP is busy I have 4 mobile phones, my Ipad and test ancient Laptop, this Laptop, and two neighbour mobile phones registered via Hotspot. In addition I have the new Groove which I am trying set up to be CPE which connects to a secn VAP and then feeds using a different inexpensive little internal user friendly WIFI access point-router-PPPoE client... all this and the Metal is still going strong at Uptime 3d 11 h with lots of changes as I try to learn how to get PPPoE to pass traffic in spite of being pretty new at this MikroTik stuff. (Hot Spot + PPPoE + two internet connections maybe is complicating? The customer gets DNS but not anything else... yet... anyone want to help?)...

I would not yet trust that newer versions of routerOS are OK until we have been told they actually found and fixed their issue.

WoZeR, I could post the version I have, RouterOS v6.18rc4 built at 13h58, to a private FTP site if you did not happen to grab it and they happened to have changed it.

Thanks Lansman but that's the current version I'm testing right this moment (routeros-mipsbe-6.18rc4.npk).

It's looking better, the device is now usable but there are some stability issues to work out. I do agree with you about waiting for Mikrotik to release an announcement that it is fixed but they do need us to test every possible scenario and let them know if it works or not here. The only thing I'm worry about if they are actually reading our post. I'm upset about the very poor support I'm receiving. For some reason Mikrotik thinks it's best to keep their customers in the dark and ignore their support tickets. At least someone there is attempting to fix this issue.

Hello WoZeR and others. It Works!

The Metal has now been up more than 7 days without failure. The very marginal 1.1 km very not line of sight link goes from my barn to my friends home .. a friend who has late stage cancer, little mobile phone coverage... so I wanted to deliver some sort of connectivity... that link has been up a total of two days with no failure. The link is a poor signal with a lot of multipath as it scatters in trees and is blocked by two buildings. I would not call it professional. But it works well enough. The signal level is less than expected. I assume that is due to OFDM seeing multipath as a contributor. A high gain antenna, such as that used at the home sees less mutipath, thus less signal.

The other end is a Groove reporting ~ -80 to -83 dBm, the Metal reports -89 to -85. Omni on the Metal at ~ 8 dBi, 24 dBi connected to the Groove. The Groove goes to a TP-Link which dials into the Metal using PPPoE and then, being inside the home, delivers a nice wi-fi signal to the user. As other users of my Metal use cell phones, a lot of the barn received signal is very low. One is usually at -90 to -89 dBm, and is often deregistered and reregistered. So the Metal has a lot of work to do.

Now that it has been up a week, I declare it to be GOOD!

Yes I agree, it's fixed.

Hi WoZer. You said, "they do need us to test every possible scenario and let them know if it works...". True. But once we identify a repeatable problem, we should offer or support should ask for a copy of our backup files so that they can test to see if it is repeatable, and if it is to let us know, then they can find a resolution in house and post their progress in the research. I did offer MikroTik support my back up file. Had they taken it and tested using it, I would not have spent 100 hours trying to solve this problem, and other users would not have had the problems mentioned in this forum thread. Support has the spare equipment to run s similar set up. Then determine what should be done.

They are at fault for not asking for our back up files and replicating the probem in house.

(I am happy they did fix it though)

We recently purchased 10 Mikrotik Metal 2SHPn routers and are having many problems with every unit. I am attaching supout files for both radios. As you will see, both units have firmware version 6.17. The basic configuration of each radio is as follows:

Radio #1:
- Mode: Bridge (with ether1 and wlan1 added as ports)
- Wireless Mode: AP Bridge
- WDS Mode: Dynamic
- WDS Default Bridge: bridge1
- Band: 2GHz Only N
- Channel Width: 20/40MHz HT-Above
- Nstreme Enabled with Framer Policy set to Best Fit and Framer Limit set to default value
- Distance set to “Indoors” since actual link distance is only around 20 feet

Radio #2:
- Mode: Bridge (with ether1 and wlan1 added as ports)
- Wireless Mode: Station WDS
- Band: 2GHz Only N
- Channel Width: 20/40MHz HT-Above
- Nstreme Enabled
- Distance set to “Indoors” since actual link distance is only around 20 feet

It should be noted that the testing described below is done in a closed loop test setup in that antennas are NOT used. The two radios are connected together using 25 feet of coaxial cable plus a 90dB coaxial attenuator to reduce signal level down to desired level. This setup ensures no wireless interference from other transmitters and virtually perfect CCQ. This setup I mention has been used for more than 10 years to test many different wireless routers that we have sold including many built with Mikrotik router boards and radio cards. This setup has worked flawlessly so please do not question the integrity or validity of this setup. Also, to verify this setup is still functional, we have tried using other wireless routers we have built using the Mikrotik router boards and radio cards (versus the Metal 2SHPn) and the problem described below does not exist. In that scenario, the only thing we are swapping out if the physical wireless radios themselves. All other test fixture components including POE’s, POE cables, coaxial cables, attenuators are all unchanged. Finally, I should also say that we have not seen this problem at all on the Mikrotik Metal 5SHPn only the 2SHPn. However the problem appears to be on every unit we test. The problem also appears to be aggravated when using Nstreme. I don’t understand this because again, we do not see this problem when using say RB433AH router board with R52Hn radio card.

I also want to say that I have reloaded flash with firmware version 6.17 using Netinstall so everything is reset to factory default from the get go.

The physical test setup, from one end to the other is as follows:

“Windows XP PC -> Ethernet Cable -> POE -> Metal 2SHPn -> 90dB Attenuator -> 25 Foot Coaxial Cable -> Metal 2SHPn -> POE -> Ethernet Cable -> Windows XP PC”

When I mention below passing data across the link, this is done using “iperf” on both PC’s, which is completely independent of the network devices between the two PC’s.

The problem we are having is this:
1) Wireless link is established very easily and will stay up for indefinitely until I start passing data across the link.
2) Once I start passing data across the link, the link will drop out completely within 2 minutes and sometimes never come back up until I reboot. Other times, the link stops passing data and the Wireless Registration table shows link is still up but last activity continues to increase indefinitely.
3) While the link is running data, the Tx and Rx Signal Levels are -57/-57 and Tx/Rx CCQ is 96/100.

Disabling nstreme eliminates this problem. I have just sent email to support@mikrotik.com. Given how long this problem has existed, based on this thread, I would certainly hope by now that Mikrotik knows about this problem and either has a fix or admits theirs and issue.

--- UPDATE ----
I just downloaded and installed 6.18 on both radios and ran the test again. Seems more stable but still have drop outs and occasionally the link freezes and never comes back up until units are rebooted as reported by others in this forum also.

There is a test 3.17.2-pcie-Metal-2SHPn firmware for these devices, but I don't know if it is included in the latest v6.18 package.
I got it from support for testing purposes. I haven't done stress tests, but I am using that firmware on some devices and I have 16 days uptime without problems.

Just another quick update. I sent a long email to Mikrotik technical support regarding the problems I was seeing and they sent me back an email with a link to a firmware patch (ar7240_3_17_2_Metal_2SHPn.fwf) specifically for the 2SHPn as the previous post mentioned. Per their instructions, I have applied this patch to all of my radios and also did a reset on the wireless interface of each radio. I have two identical test setup both running this patch and links are very stable now. I did have a couple of short wireless disconnects on one of my two test setups in the beginning but none for several hours now. Keep your fingers crossed that this problem, after a year and a half according to this thread, is now finally fixed!

@MSSFL

Do you have a link to this firmware file?

Hi im new here (sorry my bad english).

I have a metal 2shpn too and i got the same problem but i can middle fix it with this config:

- Cpu speed at 250mhz (System>Routerboard>Settings apply the cpu speed and reboot)
- Distance: dynamic
- Wireless band: 2.4ghz only B ( i got really bad connection with G and N )
- Manual txpower at 32dBm
- Country: no_country_set
- Antenna gain: 15dBi (tp link 2415d)

It really stable but got high pings like 500ms, 700ms, 350ms, but clients can surf on internet no suffer disconnection im 9km away from metal. Hope this help someone while Mikrotik team fix this problem.

P.D: @MSSFL, @CyberTod can you guys put the firmware link i wanna test it, or @normis put link here for everyone please

Write me an email : (my username) at vestel.bg and i will send it to you.

Your high ping could be due to using only 'b' rates. If usage is high 11mbit/s wireless rates is just slow and will be filled.

Write me an email : (my username) at vestel.bg and i will send it to you.

Your high ping could be due to using only 'b' rates. If usage is high 11mbit/s wireless rates is just slow and will be filled.

Done

i have upgraded to "routeros-mipsbe-6.21rc3.npk", but problem remain exist. Can anyone please give me the link of "firmware patch (ar7240_3_17_2_Metal_2SHPn.fwf)" ??????

It would greatly help if MT makes this file ar7240_3_17_2_Metal_2SHPn.fwf available somewhere for download if it really is the solution to years long problem with Metal2SHPn. There should also be an explanation or link on how to upgrade it, etc. I also have the same problem with Metal2 and hope this will fix it.

I've tossed all my metalSHPn2's in the garbage.

I would like to pay the shipping if you had sent them to me.

I have upgraded the firmware patch to (ar7240_3_17_2_Metal_2SHPn.fwf), but problem has not solved yet.

Can we replace mikrotik with ddwrt in this device for test purpose???

I have the same problem with the new RouterOS 6.22 and firmware 3.19.

Good night,

I present this problem for a year and had to restart the Metal when he stopped sending signal to customers, sometimes up to 3 times per day, this had me crazy. One day I miss definitely send it off and lit forcing with cable, and nothing (Prior to that he had already changed the POE for Ubiquiti 24v, in case there was trouble voltage as the cable is Ubiquiti 40 Meters long and neither was the solution) after that night stop sending him under the antenna and spend all day trying it to catch from 7 AM to 2 PM, until the Netinstall Install the OS program from 0 by pressing the reset button for 15 seconds erase all data on it update the firmware to 3.19 reconfigure it got all customers EUREKA and the problem was solved, no thing happened the OFF SIGNAL now come to a conclusion: Mikrotik Metal you can not update, now I explain why 01 OS RouterBOARS saves the previous dirty data and creates a conflict for subsequent updates 02 OS is highly generic for all RouterBOARS consequently resulting in poor data management hardware or device drivers in this case Mikrotik Metal 03 garbage or temporary stays in the same partition occupying living space of the equipment operation and 04 a clean installation improves the use of RAM and CPU that makes the Mikrotik Metal stop working effectively causing it not fluent in terms of data delivery, I have made this device is more efficient than any Ubiquiti with resource consumption lower himself and very fast client-server performance this makes customers feel more satisfied than before, is more compatible with multiple receiving devices before offering a better service, now the same to have a ping failure it gives a very efficient without programming anything additional automatic restart, much lower temperature, the use low CPU and RAM too low, I am currently using version 5.25 and 3.19 Firmware without fail for almost a month solved the problem with a very high customer-company guarantee satisfaction ago.

i have upgraded to v 6.22 also firmware in my Metal 2Shpn. After up gradation, last few days i have changes all possible parameters but problem has not solved.

Smartphones, Wifi Usb Dongles & Ip cameras are connected and works within 20 Meters distance. Above 20 meters these devices are connected but there is no data transfer possible also getting packet loss.

Below is my Routerboard & wireless interface details :


MikroTik RouterOS 6.22 (c) 1999-2014 http://www.mikrotik.c

[admin@MikroTik] > /system routerboard print
routerboard: yes
model: Metal 2SHPn
current-firmware: 3.19
upgrade-firmware: 3.19
[admin@MikroTik] >



/interface wireless
set [ find default-name=wlan1 ] band=2ghz-b/g/n channel-width=20/40mhz-eC \
disabled=no distance=indoors frequency=2432 l2mtu=1600 mode=ap-bridge \
ssid=raktim tx-power=25 tx-power-mode=card-rates
/interface wireless security-profiles
set [ find default=yes ] authentication-types=wpa-psk,wpa2-psk eap-methods="" \
group-key-update=1h mode=dynamic-keys wpa-pre-shared-key=today123 \
wpa2-pre-shared-key=today123
[admin@MikroTik] >

Hello all

did they solved Metal 2sHPn issue ?

I want to buy 01 Metal 2 sHPn for Access Point installation in our residential city (I want to cover ~500 meters around AP)
is this possible with only one ?

Thought i buyed something good, but never have worked with a crappy device like this..

lot of hassling around to get it work, and after a wile stops working, it blocks and have to go up in the tower to get it out and to load new firmware on the pc.

Very unhandy software to make a quick setup, many things don t work properly ...

Send it back today and stay to ubiquiti bullets.


Cant make it better than it is, losiento

I don't have access to my Metal-2SHPn to configure it. I have used Eth1 Mac and Wlan Mac address but non change. I have trying to reset with buton. No change.
There are others solutions ?

Is this still an issue with the 2.4GHz MikroTik metal?

If it's still an issue, any similar model that has no problems?
I really don't want to use something else than RouterOS.

OLA JA TENTOU MODIFICAR O PARAMBLE MODE : COLOCA ELE EM BOT EM VEZ DE LONG

Does any one of you have more than one Metal 2SHPn? Do all of your metals have this issue, or only some? Can you send the defective units directly to MikroTik for inspection? Please contact support@mikrotik.com to arrange this

I have 4 new ones that I tried, and the users have little transfer and they connect badly being only 100 meters away