Is this a bug in MT or with SR-2 card where after a few minutes or hour all of the client signal levels report in the positive 100's and they can't pass traffic?
I've seen this before but can't remember what the cause was.
This is MT 2.9.23 on RB500 series with one SR-2 card.
Any ideas?
You'll probably need to configure the TX power as posted here, I had problems with the SR2 until I did that. I really wish they'd put something in the manual about it, or sticky that thread. SOMETHING so that this question won't keep getting posted again and again.
-
-
wildbill442
Forum Guru
- Posts: 1055
- Joined:
- Location: Sacramento, CA
I think I have a bad batch of SR-2 cards then.
This is happening on almost an hourly basis now.
It happens in RB500, RB230, MT versions 2.9.23 and 2.9.7.
Tried several different settings on these cards and this still happens and one card just seems to be fried because the power levels are low.
I'm switching back to Prism 200mw cards.
This is happening on almost an hourly basis now.
It happens in RB500, RB230, MT versions 2.9.23 and 2.9.7.
Tried several different settings on these cards and this still happens and one card just seems to be fried because the power levels are low.
I'm switching back to Prism 200mw cards.
Well, our Prism cards are PCMCIA so we would have had to keep them in our older RB230 units.
Instead we replaced the apparently faulty SR-2 cards with some EM9 cards we had lying around from Mikrotik.
Now the signals are very good and stable.
No magical signal levels in the positive 100's and no rebooting every half hour (so far anyway).
Instead we replaced the apparently faulty SR-2 cards with some EM9 cards we had lying around from Mikrotik.
Now the signals are very good and stable.
No magical signal levels in the positive 100's and no rebooting every half hour (so far anyway).
The positive 100's appear due to the binaric nature of a byte. If you have ever programmed in Assembler, you understand what I am saying.
Usually, 8-bit sequence declared as type:byte allows numbers from 0 to 255. 8-bit type:small_integer stands for -127..127 and the very first bit of the byte is a sign-bit. If you decrease the negative number by 1 consequently, after -127 you will be getting bit inversion in the very first bit and the resulting number will be in positive 100's. So, it is the bug in the RouterOS, when signal gets too weak, it jumps to positive value.
Usually, 8-bit sequence declared as type:byte allows numbers from 0 to 255. 8-bit type:small_integer stands for -127..127 and the very first bit of the byte is a sign-bit. If you decrease the negative number by 1 consequently, after -127 you will be getting bit inversion in the very first bit and the resulting number will be in positive 100's. So, it is the bug in the RouterOS, when signal gets too weak, it jumps to positive value.
Who is online
Users browsing this forum: No registered users and 2 guests