New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
journal/syslog floods, exhausting disk space #588
Comments
1 million, huh? That is... excessive! Wow. Those areas of the code have been reworked quite a bit since v24 was released. (Ages ago, in gnome-shell-extension years, since it'll be 3 months old on Friday.) With a v25 release beginning to take shape on the horizon, I was wondering if you'd be willing to try installing the v25-rc3 release candidate to see if it clears the problem up? You can find "manual" upgrade instructions for installation from zipfile on the Wiki's "Installation" page. |
Thanks, I will try this at the next opportunity and reply with any results. |
Hi, just to report I had the same experience today. My logs grew to over 50GB then my machine pretty much just died (also on V24). The messages were mostly:
|
Now that |
I've had no problems since trying the beta, and now the latest release. |
Thanks @JamesSwift , glad to hear it! I'm going to close this, if this same problem occurs (for anyone) with v26, please open a new issue with relevant log messages. |
GSConnect V37. Syslog about 50 gigabytes!
|
Did you recently upgrade from Ubuntu 18.04? If so you may have to start with a fresh configuration for GSConnect. See this page in the wiki. |
Seeing this right now under Manjaro. I have |
I had this problem today on Ubuntu 20.04 and it had filled up my syslog with 331gb. Full disks on linux isnt pretty and I thought my machine was dead since X wouldnt start. Had to ctrl-alt-F2 to TTY to find and delete the syslog.1 -file in order to get my (x) to boot properly. I was about to buy a new computer, as I assumed my 4yo computer had died. If im correct with posting this experience on this issue, which maybe not correct...? These are the two lines repeating in my logs (331gb): ** |
Hello, After debugging, turned out that snapd was the cause root, apparently it installed multiple gnome related things by default, and probably did an auto update.. → Don't have an other explanation, the system was installed more than 6 months ago, and no one touched it since. After disabling all snap related services and emptying the syslog, things are back to normal. Distro/Release: Ubuntu 20.04 |
@finnjohnsen could you post some lines of the actual log? I'm especially interested in what is the interval at which these repeat. Are you sure that GSConnect was causing the log spam, or could it have been something else? To the original post, one reason the journal is smaller is that it has actual settings for max storage usage as well as limits of how much at a minimum to try and keep free. |
Just had the same problem - Ubuntu 20.04.1 LTS I dual use the computer to 1) run my TV with gnome and 2) run a nextcloud server. I can see from Prometheus that my disk started filling at ~6.30am today and I caught it after it consumed ~135gig at 10.30am - a few hundred meg away from zero disk space. I've already truncated syslog to keep the server alive, but I've captured the following from the console as I investigated: Syslog: Top: top - 10:09:59 up 7 days, 12:44, 2 users, load average: 3.45, 2.81, 2.56 ps: chris@viper:~$ ps ax | grep gsconnect Resolution: After killing gjs and performing no other action I can't see any negative impact on the system and everything seems to be running as normal. |
I'm sure the daemon probably shouldn't crash at all, but that doesn't look like enough volume to fill >100GB in a matter of a few hours. I'd very much like to help, but so far I have no good idea where to look. |
I'm also seeing this error, it's intermittent but when it happens my gnome-shell completely freezes. I am running gsconnect, I'll disable it for a little while and see if the crashing goes away. |
In my case I don't think it was gsconnect, after more debugging I disabled hardware acceleration in my browser (Brave) and I haven't had a crash since... |
Ubuntu 21.04 & GSConnect 45. I had the same issue as the OP (my syslog grew as much as 900 Gb). I needed to uninstall the extension, reboot my computer and install GSConnect again. After that all seems to work ok. |
Weirdly, after a few months of this not happening, I'm now experiencing regular desktop freezes with this same message again. Will try removing GSConnect and see what happens... |
I've just experienced the same problem. My entire root partition has been filled by
Logs fragment:
Also, this warning shows up randomly:
|
Describe the bug
Having GSConnect installed and enabled periodically leads to a full disk. As a result, rsyslogd takes up 100% CPU, lots of fan noise and heat, etc.
Inspecting the journal (
$ journalctl
) yields more than 1 million repetitions of messages like:…and the same four messages in /var/log/syslog. /var/log/syslog ends up filling the disk (at >10 GB); however the journal (/var/log/journal/*) is also >1 GB, probably only smaller because it's stored as binary rather than plain text.
Steps To Reproduce:
Expected behavior
Logs are not filled with GSConnect-related error messages.
Disk space is not exhausted.
Support Log
The GSConnect settings dialog cannot be opened while the logs are flooding, and so a support log cannot be generated. It is necessary to kill the gjs process (e.g. 7642 in the above messages), after which it seems to be automatically restarted, and the dialog can be opened.
System Details (please complete the following information):
GSConnect environment (if applicable):
The text was updated successfully, but these errors were encountered: