Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
How to Troubleshoot (Seemingly) Random Freeze?
View unanswered posts
View posts from last 24 hours

 
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware
View previous topic :: View next topic  
Author Message
statikregimen
Apprentice
Apprentice


Joined: 16 Jul 2011
Posts: 173
Location: USA/Michigan

PostPosted: Sat Sep 23, 2017 5:20 pm    Post subject: How to Troubleshoot (Seemingly) Random Freeze? Reply with quote

Hi,

Well, this one is tricky as I cannot willfully reproduce it. The issue is what the subject says. Usually, it just wont wake up from sleep, but this computer idles a lot. I have also had it freeze during normal use (in which case, it's sudden, without any warning and doesn't seem dependent on task or system load).

The machine is a Lenovo Yoga 2 11" laptop.

Here is the last of syslog from the most recent freeze, when I tried to wake the system up and it was dead:

Code:
Sep 23 10:59:01 adaline CROND[25260]: (root) CMD (rm -f /var/spool/cron/lastrun/cron.hourly)
Sep 23 11:00:01 adaline CROND[25759]: (root) CMD (test -x /usr/sbin/run-crons && /usr/sbin/run-crons)
Sep 23 11:10:01 adaline CROND[30515]: (root) CMD (test -x /usr/sbin/run-crons && /usr/sbin/run-crons)
Sep 23 11:20:01 adaline CROND[2721]: (root) CMD (test -x /usr/sbin/run-crons && /usr/sbin/run-crons)
Sep 23 12:40:15 adaline syslogd 1.5.1: restart.


The last line, I'm not sure why it says "restart" b/c it was a hard power off, but that is probably just semantics.

I suppose it could be one of the 2 cron scripts but I doubt it....

Code:

# ls /etc/cron*
/etc/cron.deny  /etc/crontab

/etc/cron.d:

/etc/cron.daily:
logrotate  man-db

/etc/cron.hourly:

/etc/cron.monthly:

/etc/cron.weekly:


I do not believe it to be hardware failure, but it's not out of the question. It's an older, lower end laptop. However, I do not think this issue started until I switched from 4.8 hardened kernel to gentoo-sources. During the change, I added additional processor governers and am using On Demand (vs. Performance before), as well as S3/Suspend to RAM. I tried disabling S3 to no avail. I have not tried changing governor, but I really need that to work b/c this machine is passively cooled. Nevertheless, I'll try setting it back to performance only.

Any further troubleshooting strategy is beyond my experience right now, since I cannot reproduce it at will, but it does seem to happen at least once or twice a day, so will be easy to know when it's solved, at least...

Thanks for looking and for any help!

EDIT:
Last few lines from syslog on latest freeze (was just chatting on steam at the time):

Code:
Sep 23 20:20:01 adaline CROND[800]: (root) CMD (test -x /usr/sbin/run-crons && /usr/sbin/run-crons)
Sep 23 20:20:34 adaline dhcpcd[3290]: wlp1s0: dhcp_sendpacket: Operation not permitted
Sep 23 20:21:38 adaline dhcpcd[3290]: wlp1s0: dhcp_sendpacket: Operation not permitted
Sep 23 20:23:37 adaline syslogd 1.5.1: restart.


So no real help there, other than this time I was present and using it and those cron jobs are tighter in time to the crash, but still I'm not convinced that has anything to do with it.


Last edited by statikregimen on Mon Sep 25, 2017 10:51 pm; edited 1 time in total
Back to top
View user's profile Send private message
pjp
Administrator
Administrator


Joined: 16 Apr 2002
Posts: 20067

PostPosted: Sun Sep 24, 2017 5:44 am    Post subject: Reply with quote

To clarify, by freeze you mean that it permanently locks up until you force the power off, correct?

How much time passed from those messages to the crash? It doesn't seem likely to be related, but you could trun them off temporarily as well.

Normally hardware is the first culprit to eliminate. Often power supplies and memory. But since you changed kernels, those changes are also high on the list.

If you still have the hardened kernel available, I'd switch back to that to verify if the problem goes away. If it goes away, then kernel changes are more likely the issue. But it still might be a good idea to test the memory.

If you switch back to hardened and it seems fine after a while, then I'd save that configuration so you can revert back if necessary. I'd then try to use a gentoo-sources kernel of the same version, without making any new changes (hibernate, governor, etc.). Try to match it as closely to the good kernel as possible.

If you can't find a close version of gentoo-sources, then I'd use the hardened kernel and disable the hardened features in an attempt to make it closer to a gentoo-sources kernel. It might then be easier to migrate from that "unhardened" kernel.

Also, /usr/src/linux/scripts/diffconfig can help you compare kernel config files.

And one more thing. Since it is passively cooled, it may be overheating. Does the system have any temperature sensors which can be monitored?
_________________
Quis separabit? Quo animo?
Back to top
View user's profile Send private message
bunder
Bodhisattva
Bodhisattva


Joined: 10 Apr 2004
Posts: 5934

PostPosted: Mon Sep 25, 2017 5:55 am    Post subject: Reply with quote

Try enabling magic sysrq and seeing if the reboot button works. If it does, then its not a complete hardlock. You might be able to generate a stack trace. I remember ages ago when video drivers weren't so great, that x11/xorg would freeze and give the appearance of a hardlock.
Back to top
View user's profile Send private message
statikregimen
Apprentice
Apprentice


Joined: 16 Jul 2011
Posts: 173
Location: USA/Michigan

PostPosted: Mon Sep 25, 2017 10:50 pm    Post subject: Reply with quote

Thanks very much for the replies! I will try these suggestions ASAP.

As for memory, I feel like it's soldered on to the mainboard on this machine, but I can at least run memtest86 or so. Unfortuately for me, this is pretty much a disposable computer, tho it is nice so I hope it's not hardware!!

Again, I thank you. I shall report back my findings...hopefully within the next 48hrs from this writing :)

EDIT: So it's definitely not overheating. I'm pushing it hard right now, and not only is it running really hot, but it's running smooth. My current load average is >6 on the minute.The major task running, is an emerge @world update. Yet still, my GUI apps are responsive as ever. I mean, there are a few lags here and there, but truly impressive for a Baytrail. No other OS/distro has performed so well on this machine to date. Heavy load, but still feels like it's prioritizing me. Battery life has improved, too.

EDIT2: The load condition has been sustained for at least an hour now....no freeze so far.

EDIT3: It did complete the update after a great deal of time (hours - I went to bed eventually), and it stayed alive over night. I'm not sure I've had a freeze since then, so currently, I am monitoring the situation.

EDIT4: It froze again, eventually... Should have some time to apply the provided strategies this weekend.
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware All times are GMT
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum