NETGEAR is aware of a growing number of phone and online scams. To learn how to stay safe click here.
Forum Discussion
Solved
New - RBRE960 / RBSE960 Firmware Version 7.2.6.31 Released
Enhancements: 03/22/2024
(For all regions except Europe)
Improves the stability of WiFi device connections during roaming
This firmware addresses security vulnerabilities. For more informatio...
I have noticed issues that only seem to crop up after days of running - as few as 3-4 days, as many as 22 days. They are router or satellite restarts, memory growth in satellites to the point of causing a reboot and disconnect of all devices. I have seen WiFi disconnect and stop working until the router was rebooted. Have seen IPv6 stop working until it was re-enabled (likely a Comcast-NG interaction issue). I have seen satellites lose their backhaul connection with the router.
I believe at least the memory creep issues are related to using 2 satellites on Ethernet backhaul, just a guess. It could also be related to higher speed networks, mine is a 2G/200MB service. Maybe it's a combination of different factors that make the perfect storm that problems can appear? The % of Netgear customers with 2G and multiple Ethernet-wired backhaul satellites is small, I suspect. Plus, if there are some bugs that pop up after a number of days of running, NG has said that this kind of testing is very hard for them to do.
I just went back to v.21 to see how it runs. If v.31 is working well, great - if not, some caution could be warranted to try V.21 before spending lots of time trying to find a problem.
I updated tonight. All good so far.
I have been doing some testing and analysis with the 7.2.6.31 FW update on a RBR/RBS 960 network. It's chronological update through the day.
Here is my update from today, some interesting data to share on my 7.2.6.31 testing. Note - IPv6 IS ENABLED (I have had trouble with that, Comcast ISP, in the past)
1) This morning, the 3rd floor RBS was showing HC of blank. It started last night late. By the time I went to bed, it had disconnected any devices connected to it.
2) At the time when hop count went from 1 to blank, meaning devices would no longer connect to it via WiFi - CPU shot up to 26%. By this morning CPU was 29%.
3) With IPv6 enabled, the Router memory climbed to 580MB. Without IPv6 it's normally 360-390MB. For the last 2+ hours it has stayed up/down in the 576-580MB range.
4) I power cycled the 3rd floor RBS two hours ago, everything looks stable now. Devices are connected to the 3rd floor RBS again.
5) Here's a snapshot of each at this time, you can see when the one RBS went from hop count 1 to blank - and its CPU shot up:
Router:
Time: 03/24/2024, 08:29:30 CPU %: 11.81% Memory Usage: 580 MB/818 MB Uptime: 0 days 14:02:39
Time: 03/24/2024, 08:39:28 CPU %: 11.81% Memory Usage: 580 MB/818 MB Uptime: 0 days 14:12:37
Time: 03/24/2024, 08:49:27 CPU %: 11.80% Memory Usage: 576 MB/818 MB Uptime: 0 days 14:22:36
Time: 03/24/2024, 08:59:25 CPU %: 11.79% Memory Usage: 577 MB/818 MB Uptime: 0 days 14:32:34
Time: 03/24/2024, 09:09:23 CPU %: 11.78% Memory Usage: 577 MB/818 MB Uptime: 0 days 14:42:32
Time: 03/24/2024, 09:19:22 CPU %: 11.75% Memory Usage: 579 MB/818 MB Uptime: 0 days 14:52:31
RBS that just went to HC blank last night:
Time: 03/24/2024, 06:48:09 CPU %: 29.46% Memory Usage: 394 MB/812 MB Uptime: 0 days 12:8:40 HopCount:
Time: 03/24/2024, 06:58:11 CPU %: 29.49% Memory Usage: 377 MB/812 MB Uptime: 0 days 12:18:42 HopCount:
Time: 03/24/2024, 07:08:13 CPU %: 29.51% Memory Usage: 377 MB/812 MB Uptime: 0 days 12:28:44 HopCount:
Time: 03/24/2024, 07:18:15 CPU %: 29.52% Memory Usage: 378 MB/812 MB Uptime: 0 days 12:38:46 HopCount:
Time: 03/24/2024, 07:28:17 CPU %: 29.54% Memory Usage: 378 MB/812 MB Uptime: 0 days 12:48:48 HopCount:
Time: 03/24/2024, 07:38:19 CPU %: 21.8% Memory Usage: 324 MB/812 MB Uptime: 0 days 0:2:19 HopCount:
Time: 03/24/2024, 07:48:21 CPU %: 14.4% Memory Usage: 343 MB/812 MB Uptime: 0 days 0:12:21 HopCount: 1
Time: 03/24/2024, 07:58:23 CPU %: 11.32% Memory Usage: 342 MB/812 MB Uptime: 0 days 0:22:23 HopCount: 1
Time: 03/24/2024, 08:08:25 CPU %: 10.15% Memory Usage: 343 MB/812 MB Uptime: 0 days 0:32:25 HopCount: 1
Time: 03/24/2024, 08:18:27 CPU %: 9.46% Memory Usage: 344 MB/812 MB Uptime: 0 days 0:42:27 HopCount: 1
Time: 03/24/2024, 08:28:29 CPU %: 9.05% Memory Usage: 344 MB/812 MB Uptime: 0 days 0:52:29 HopCount: 1
Time: 03/24/2024, 08:38:31 CPU %: 8.8% Memory Usage: 344 MB/812 MB Uptime: 0 days 1:2:31 HopCount: 1
Time: 03/24/2024, 08:48:33 CPU %: 8.63% Memory Usage: 344 MB/812 MB Uptime: 0 days 1:12:33 HopCount: 1
Time: 03/24/2024, 08:58:35 CPU %: 8.49% Memory Usage: 345 MB/812 MB Uptime: 0 days 1:22:35 HopCount: 1
Time: 03/24/2024, 09:08:37 CPU %: 8.4% Memory Usage: 345 MB/812 MB Uptime: 0 days 1:32:37 HopCount: 1
Time: 03/24/2024, 09:18:39 CPU %: 8.31% Memory Usage: 345 MB/812 MB Uptime: 0 days 1:42:39 HopCount: 1
The other RBS:
Time: 03/24/2024, 08:23:59 CPU %: 9.76% Memory Usage: 369 MB/812 MB Uptime: 0 days 13:51:4 HopCount: 1
Time: 03/24/2024, 08:34:01 CPU %: 9.75% Memory Usage: 369 MB/812 MB Uptime: 0 days 14:1:6 HopCount: 1
Time: 03/24/2024, 08:44:03 CPU %: 9.74% Memory Usage: 369 MB/812 MB Uptime: 0 days 14:11:8 HopCount: 1
Time: 03/24/2024, 08:54:05 CPU %: 9.72% Memory Usage: 370 MB/812 MB Uptime: 0 days 14:21:10 HopCount: 1
Time: 03/24/2024, 09:04:07 CPU %: 9.71% Memory Usage: 370 MB/812 MB Uptime: 0 days 14:31:12 HopCount: 1
Time: 03/24/2024, 09:14:09 CPU %: 9.7% Memory Usage: 370 MB/812 MB Uptime: 0 days 14:41:14 HopCount: 1
6) Funny how both the last two times after a power cycle of everything, within a day I had an RBS (different one) fail with hop count blank. I then went 9-10 days with no problems after that. Last time it was with IPv6 disabled, this time IPv6 is enabled.
7) I noticed one other version change on the routers debug page, from the FW upgrade to 7.2.6.31. “Seal“, whatever that is, went from 2.15.49–0 to 2.15.50–0. I wonder what that is?
😎 OK, I just noticed that IPv6 has failed. I am going to disable it and power cycle everything again. Things seem to get really unstable with IPv6 enabled here.
9) There is always so much drama on power cycles. It's like it can't handle 2 RBSs coming up near the same time. I always bring one up, wait for the blue LED, then power up the other. This time the 1st floor RBS came up Blue, as did the other; but 20 minutes in the stats showed the 1st floor RBS had hop count blank (it never did show hop count of 1), and the router did a big hop in memory from 353 to 403. The other RBS was HC 1 in 7 minutes. So I power cycled the 1st floor RBS again, the first report at 6 minutes showed all good and HC 1.
10) Ugh - 1st Floor RBS just went Hop count blank at 16 minutes powered on.
11) This time I powered off the RBS and RBR, waited 2 minutes, brought up the RBR, then waited TEN minutes for things to settle down. I just powered up the RBS. IPv6 has now been disabled. But Hop count is STILL blank - and nothing connecting to the satellite.
12) I tried power cycling again, and it never got out of hop count blank. So this time I powered off the other RBS, took the offending RBS down next to the router, let it sync wirelessly, then brought it up and powered it on. Hop count is 1 so far!
My suspicions: there are bugs with:
- Ethernet backhaul. I have had similar issues to this with the Orbi 850 series too, as well as with 7.2.6.21 and 7.2.6.31 Firmware. Another user has had them on his network. WiFi seems more stable.
- Power cycling the network seems very hit or miss. I frequently have to power cycle on of the RBS AGAIN.
- Hop Count turns off for no apparent reason - then nothing connects via WiFi. Note, anything connected direct continues to work, so the backhaul is fine (and it is reported as fine on the router admin web page)
- Last night router was in a steady memory creep - like there is a memory leak.
Thank you donawalt Great information.
Seems like IPv6 is being problematic, again. 🙄Ya, we need to get some info on Hop Count behaviors. Seems like when RBS are ethernet connected, something stops working after period of time and seems to stop the RBS from working properly.
I recommend that you open up a NG support ticket with your 960 series and put all this in a support ticket to help with letting NG know about this. Seems like this should not be happening and something in FW needs to be checked into.
