@NOW wrote:
I have all the following CPEs connected to a 2000 series AP and the all seem to be ignoring the MIR profile settings.
By seem, I mean when using the speedtest tool in the AP or in the CPE
Thoughts?
.
MIR works well for us in the lab with 4.4. FW.
Could you please send configuration files to me at fedor.trutsko(at)cambiumnetworks(dot)com?
Thank you.
@Fedor thank you for the reply!
Good Day,
Why is it that our Epmp3000 still shows online with it’s children connected in our Cloud cnMaestro when the 3000 is disconnected and sitting next to me unplugged?
@jnovak wrote:I was able to login to the force200 that was rebooting, reflash the 4.4 firmware, and it's been stable since.
All that leaves is XW elevates radios connected to this sector. All but one of them are unstable (5-6 units). All of the XMs work as expected. At least a couple of the radios reboot when attempting to log in to them remotely, one of the radios just seems to disconnect/reconnect but at least it doesn't reboot every single time.
Okay, I've got a final on this issue, except for the XW radio issue as we swapped those out quicker then I realized the below.
This isn't obvious as far as I saw. Remember EPMP3000 is limited to 5ms frame time right this moment. I don't know how many know this, especially many like me who finally deployed it now that 4.4 released and a lot of the mixed mode issues have been worked out. My problems where entirely self interference; we've probably got something like 20 APs within 6 miles of this single AP we deployed.
After moving around some channels, and changing frame timing on a couple of key APs, everything is working as expected, and from what it seems like, airtime usage is lower and bandwidth is up just a hair - just with N based CPEs alone.
Joe
Thank you Joe! We also didn't catch that 2.5ms wasn't supported and put a couple up on sites with 2000s with 2.5ms. We were just about to order more, but I guess that will have to wait.
@3-dBnetworks wrote:Thank you Joe! We also didn't catch that 2.5ms wasn't supported and put a couple up on sites with 2000s with 2.5ms. We were just about to order more, but I guess that will have to wait.
no problem! I'm not going to let it make me wait. I'm just going to be a bit more selective with my frequency re-use for the time being, I even considered pushing to 5ms network wide during the transition.
Just to chime in my 2c - the #1 compelling reason for most people to choose 2.5ms frames with ePMP1000/2000, was to lower latency. There are other reasons for some people, but lowering latency was the #1 reason. With the ePMP3000, the latency is typically lower on it's 5ms frames, then it was in ePMP1000/2000 with 2.5ms frames.
So - moving the whole network to 5ms frames isn't a horrible idea, if it allows you to deploy 3000's sooner and migrate customer's over sooner - you may find an overal latency benefit, even wtih larger frames.
EDITED: Cambium has said that 2.5ms frames are being tested now (Aug 29/19) and may appear in firmware 4.5 in the Sept/Oct 2019 timeframe.
Good input. Forgot about the decreased latency on the 3000s. You'd also gain about 20% capacity increase. For us the issue is syncing with another WISP that has some 450 APs in the same area. My understanding is that sync between ePMP and 450 is not supported with 5ms frame size.
@3-dBnetworks wrote:My understanding is that sync between ePMP and 450 is not supported with 5ms frame size.
Sync between 450 and ePMP works just fine with 5ms. I use it on all my sites, and it's supported in the co-location spreadsheet available HERE.
Thanks for clarifying Eric! I heard that from someone reputable and didn't bother to check. My bad.
any one else having lock ups on 4.4 f/w where all subsriber modules stop working till ap is rebooted ? GPS is fine. Just watching our support chat and they have rebooted about 7 to 10 access points so far this weekend where all subscriber modules drop until access point is rebooted. These are all ePMP 2k Access points
@Dave Mills wrote:any one else having lock ups on 4.4 f/w where all subsriber modules stop working till ap is rebooted ? GPS is fine. Just watching our support chat and they have rebooted about 7 to 10 access points so far this weekend where all subscriber modules drop until access point is rebooted. These are all ePMP 2k Access points
Are any of your AP's or SM's running mixed 3.5.6 and 4.4 firmware? e.g. a 3.5.6 AP with 4.4 SM's... or perhaps a 4.4 AP and mostly 4.4 SM's but with a few 3.5.6's mixed in there?
There's a bug with mixed firmware that causes the AP to drop all the SM's and requires a reboot of the AP to restore service... although it's just temporary. The fix (for the time being) is to make sure you're running all 3.5.6 or 4.4... not a mix. There's a fix hopefully coming in 4.4.1.
@Dave Mills wrote:any one else having lock ups on 4.4 f/w where all subsriber modules stop working till ap is rebooted ?
The only place where we've seen this is where we've accidently have mixed up firmwares. If we're running 4.4 with all 4.4 clients, we haven't seen it (at least in our environment we havent) and likewise if the AP is 3.5.6 and we have 100% 3.5.6 clients, we've also been find. The only 'oddness' we've seen is when we've got mixed up firmwares on the same AP/Sector - which is generally a no-no anyway.
Ok thanks all,
will look into making sure both banks are running 4.4 and all subs as well
Hello all, still having to perform lots of reboots on 4.4 f/w even after confirming all subs are running 4.4 f/w as well as both banks in AP's being on 4.4 f/w. This is with ePMP2k 5GHz Access points and ePMP1k 2.4GHz access points running all ePMP or a mix of ePMP elevate radios. Symptoms seem same 4.4 f/w, GPS ok, zero subs connected, reboot access point and subscribers come back online.
after updagrading my epmp1000 to non GPSsync 4.4 it only goes to login page but doesn't excute and redirect to login page again please help
@syed umair wrote:
after updagrading my epmp1000 to non GPSsync 4.4 it only goes to login page but doesn't excute and redirect to login page again please help
Hi,
Is it pingable?
Thank you.
What version did you upgrade from? Was it 3.5.2 by any chance? If so, you may need to delete the “site specific cookies” for that IP.
Updated first 2 APs (2000) to 4.4 and one of them failed to come back. Crew is heading out to replace the module so we can look at it, but now I'm scared to death to update anything. Was coming from 3.5.6 and updated via Maestro.
We found the AP in a 99% default state after the failed cnMeastro update. It lost all programming except the password. It took the update on bank 1 tho. So I updated bank 2 via webpage and it updated fine, reprogrammed and its back up and running. Any ideas what caused this? Cost us an hour and a half drive each direction.
