bonienl
-
Posts
10233 -
Joined
-
Last visited
-
Days Won
65
Content Type
Profiles
Forums
Downloads
Store
Gallery
Bug Reports
Documentation
Landing
Report Comments posted by bonienl
-
-
The state of the help display is kept in a cookie. Clear your browser's cache to remove cookies.
-
Somehow your syslog did not capture the call trace.
Anyway, I am quite confident that the patch introduced in kernel 5.10.27 will help solve our issue.
Once Unraid 6.9.2 is released, this can be tested.
-
13 hours ago, BasWeg said:
For me it does not work.
Correction will be in the next version 6.9.2
-
Sorry my mistake, it is under Settings.
-
There is an interesting kernel bug fix, which looks like our case https://lkml.org/lkml/2021/3/29/499
I don't know if this already available in a linux version for Unraid, perhaps @limetech can tell?
-
1
-
-
Go to
ToolsSettings -> Syslog Server -> Mirror syslog to flash = YesThis will keep a real-time copy of the syslog file on your flash device in the folder /logs.
-
3 hours ago, isrdude said:
Add me to the list of those with the issue.
Your syslog and diagnostics are taken after rebooting the system. Unfortunately the relevant information is gone once rebooted. Next time, please take diagnostics before rebooting (and/or activate flash mirroring).
-
13 minutes ago, jonathanm said:
Will it still flag mover failure messages?
I believe so, but would need an example message to confirm
-
8 hours ago, Squid said:
Nothing you can do about it, and pretty much nothing the devs can do about it either.
Actually the devs can do something about it
I made a correction to not flag mover messages.
-
1
-
-
9 minutes ago, CorneliousJD said:
I also never had call traces "build up" - it was one and done
In this case, is it the connection to the server not working anymore, or the complete server halted?
In other words local console is still working in this case?
-
If you have a way to create these call traces on demand that would be helpful (of course we need diagnostics to further investigate). I have host access enabled but don't have any of these call traces, and as such it is hard to reproduce the issue for me.
In the next Unraid version some more conntrack modules will be loaded, it would hopefully help to tackle the problem in more detail.
-
2
-
-
People, please post diagnostics.
-
16 minutes ago, Jimmy said:
.sometimes in the logs I have a Kernel Panic that appears.
I don't see the call trace specific to this issue in your log files.
-
1
-
-
3 hours ago, Jimmy said:
My web interface crash randomly and I think I did not have a static IP for my dockers. Here my logs
You have a different problem.
Your system is running out of memory and I also see disk full statuses.
Better create a report under General Support
Start your system in safe mode and post your diagnostics under that report.
-
Please attach complete diagnostics, a screenshot doesn't show the relevant information.
In case your server doesn't respond anymore, consider to activate the Mirror syslog to flash function, see Tools -> Syslog server. This would save a copy of your syslog to flash and can be retrieved afterwards.
-
Not really a bug, but operation by design.
When a check schedule is selected, the GUI ensures the information is always up to date when opening the page.
Alternatively the user can select to do manual update checks.
-
I had a single call trace yesterday evening.
It seems to me the common denominator is iptables, which crashes.
Unraid 6.9.1. is running iptables version 1.8.5, I haven't checked the version on Unraid 6.8.3.
-
I created on my test server a Windows 10 VM connected to br0 and I created a Firefox browser container on br0 (macvlan) too. The container has a static IP address, while the VM obtains its IP address by DHCP.
Next I open a browser in the VM and I open the container browser and let both run Internet speed tests (I have a 500 Mbps symmetrical Internet connection).
Initially one browser at the time and full upload and download speeds are achieved, then running tests simultaneously in both browsers, sometimes first starting the container browser and sometimes first starting the VM browser. Again speeds are obtained as expected (reduced speeds due to simultaneous testing).
During this ordeal everything stays stable and no call traces are happening.
I have docker "host access" enabled, but I don't believe this has anything to do with the issue.
In short reproducing this issue is a challenge ... (but I knew this)
Currently I have a 3 hour relaxing youtube video running in both browser, see how it goes!
Update
Both 3 hour long music videos ended without problem, all still going strong. I let it sit there for another day.
-
1
-
-
I made a correction for this.
Thanks for reporting.
-
1
-
1
-
-
I made a correction for this.
Thanks for reporting.
-
1
-
1
-
-
When people encounter issues like these, the first thing to try is starting the system in safe mode and stop (temporary) docker containers. This is to make sure that any plugins and docker are not the culprit.
It happened all too often that people run older versions of libraries (due to plugins or other customizations), which conflict with the latest versions which come with a new Unraid release.
-
A common cause for these call traces is when VMs and Docker share the same interface.
You should change the network model of the VMs to "virtio-net" to avoid VM/docker conflicts.
-
This is an issue specific to your system and doesn't sound like a bug.
You better post your questions under General Support.
To restore the previous version, go to Tools -> Update OS and click "Restore" (reboot required)
-
Unraid 6.9 supports multi-languages, and translatable text is surrounded by _()_ tags.
Your customizations to the GUI have broken the multi--language support (and perhaps more).
Please test without GUI customizations.
[6.9.1] All help banners are displayed (SOLVED)
in Stable Releases
Posted
The GUI is tested with all the main browsers: Chrome, Edge, Firefox, Safari and Brave.
I have never encountered the situation you describe. Sometimes browsers screw up things and the usual approach is to clear the browser's cache, but if you find a way to reproduce your issue, we can always look into it.