SOHO: Lockups & Reboots


#42

Suspect this can be related to WIFI interference issue. Do you have WIFI analyzer that can use to scan surrounding channels used when you found the connection intermittent issue ?


#43

@PeteTrax if I may ask, do you have a cellular modem on the USB port of the SOHO to use as a fallback if the main WAN port loses connection? We have been trying to identify if the random crashes are linked to having a cellular modem on the USB port or not (we have one). If you have one then it proves nothing, but if you don’t have one then it indicates the random crashes are probably not related. Thanks for your input :slight_smile:


#44

@peparn no I do not have a cellular modem on the USB port. Port 4 has a line cord I use to connect to my Mac when I want to connect to the router without WiFi. Port three has a line cord connected to an Arlo base station which is used for outbound traffic from the base station. All other ports are clear.

Bob


#45

I am sorry sitlongs, I do not.

Bob


#46

Yes @peparn I would get lock ups about every three days or so following the 7.1.1 update. Prior to the update the lockup’s would not occur for 30 days or so. Since 7.1.1 and the more frequent lockups I resorted to doing daily software reboots. I thought I was in the clear till the most recent lockup. So now I am experimenting with hardware reboots by cycling power daily. I want to see if it eliminates the issue.

Bob


#47

Indeed, very similar to what we have been experiencing with 7.1.0 and 7.1.1. I believe the problem that remains with 7.1.1s064 is the one that is hard to grasp given its random nature and sometime long time between occurrences. I am almost certain this issue was present in 7.1.0 already and may have been exacerbated in 7.1.1 in some cases, which may explain why it seems more frequent on 7.1.1. I think I mentioned this earlier but the Watchdog could free you from having to do manual power cycling daily, and just automatically restart the SOHO when it crashes.


#48

Thanks for sharing this valuable information @PeteTrax.
@sitloongs I think if their random crashes are the same than those we are having, then it rules out the USB WAN cell modem failover as the source of the crashes. It is a good datapoint I believe. It is a hard problem to sort out… :sweat: :sweat: :sweat:


#49

@sitloongs would you recommend I turn on watchdog instead doing a manual power cycle each day?

Thanks

Bob


#50

Ok after doing some testing with power cycle reboots the router still failed within 11 hours following the power cycling at 0800.

Some things for the Peplink team to consider.

  1. 7.1.1 seems to fail more quickly than 7.1.0, so what changed?
  2. Prior to the 7.1.1 upgrade the router would fail for me, but it would fail after about 30-45 days. However, one thing of note, I was not living in the house streaming and using numerous WiFi devices. The only things that were up and running were the SimpliSafe security system w/ camera and the Netgear Arlo cameras.
  3. Since occupying the house full time we have upgraded the router to 7.1.1, added more WiFi devices (2-iPhones, 2-iPads, 1-Samsung SmartTV, and two ROKU sticks.
  4. I am wondering if with the increased downloading volume due to streaming, there isn’t a processor/context switching issue that is causing the system to hang. The system hung around 2308 on November 3, 2018 after having been power cycled at 0800 the same day. I have included a link to volumetrics by date to help with the analysis.

Bob


#51

@PeteTrax

Thank you for the Sharing. We are working on the issue and will keep you update for the findings.

You can turn on Watchdog service to auto reboot the router, this will help to reduce the manual work.


#52

We’ve also have received reports of numerous ‘new’ deployments of SOHO devices experiencing the same behavior. All are on the latest firmware, however we replaced one customer with another SOHO and the issue ceased. I don’t see a difference in the hardware version (HW1) and we backed up the previous router and restored it onto the replacement model.

We have other clients that we’ve recently moved to a reboot schedule.

The engineering team has been requesting pictures of the SOHO, I assume there is a question about interference or heat which a picture would help with.

The number of devices/users varys so I can’t attribute it to that. One site has two voip phones only and has nothing but issues. We have deployed SOHOs which are pushing the boundaries against the recommended specs, but are having no issues.


#53

@jflanigan

Thank you for the sharing.

Most of the users posted here have opened a ticket for us to investigate and issue is solved using firmware 7.1.1s064 (SOHO MK3 - Memory Leak issue)

You have reported 3 devices having issue.

  • 1 device after replace with other new device SOHO MK have no issue (Running more than 20days ++)

  • The other 2 devices, from the logs shown that is the WAN connection having issue.

  • WAN health check failed - WAN disconnected (WAN failed DNS test)
  • Physical connection issue - WAN: WAN disconnected (No cable detected)

#54

@sitloongs

Thanks, is there a way to determine if our intermittent issues are related to the memory leak issue? Is there a particular hardware version or scenario that causes the memory leak? As I mentioned we have 50+ SOHO deployments and it seems only the most recent have exhibited issues. This ticket has been 20++ days old and the one issue being resolved by mailing out a new device was something we did because were weren’t getting closer to a resolution.


#55

@jflanigan,

You can arrange to upgrade the devices that having issue using firmware version 7.1.1s064 and this should fix the memory leak issue.

The others two devices given in the ticket, we found the issue are related to WAN connection disconnected issue. 1 of the device uptime is already reaching to 27 days and having WAN connection issue below:

  • WAN health check failed - WAN disconnected (WAN failed DNS test)
  • Physical connection issue - WAN: WAN disconnected (No cable detected)

We need your help to isolate the issue, that mostly related to the WAN connection issue (ISP) more than the issue reported for this forum post. Would you please kindly help to check on the WAN connection ?

I have posted more devices logs in the support ticket, you can further check on that regarding to the WAN disconnection problem.


#56

Up for 76 days… then came in today and WiFi down 2.4 & 5GHz. Cable connection was fine. Reboot through console didn’t work. Power-off, power-on solves it. Interestingly enough a recent TLS connection issue didn’t happen when we fell back to wired… More on that later


#57

Looks like this is the same problem I am running into. SOHO MKR on the latest firmware. Connected to wireless hot spot.

The SOHO is randomly locking up for 15 or so minutes then it starts working again. During this time it can’t be accessed via hardline.

The hot spot remains active during this lockup and devices connected outside of the SOHO have access to the internet.

Devices connected to the SOHO (during locked up state) reflect a “connected / no internet” msg.

I would connect the hot spot to the SOHO directly but Peplink doesn’t seem to have drivers for the Jetpack 8800 hot spot to connect to the SOHO.

I do have a support ticket started just haven’t heard anything on it.

Matt


#58

Look like this is related to WIFI AP/WIFI WAN connection problem, do you manage to download the diagnostic reports ? Please submit a ticket and attach the diagnostic report for support team to check.


#59

@pcmstock

Support team will followup with you via the support ticket. Your reported issue seem different compare to the issue mention by @SOHO as the problem is happen after running 76 days and only the WIFI connection having problem.


#60

What is the latest firmware version for this unit that works ?


#61

Please go for the latest firmware version.