Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.debian.user > #167081 > unrolled thread
| Started by | Michael Lange <klappnase@freenet.de> |
|---|---|
| First post | 2016-04-25 17:40 +0200 |
| Last post | 2016-04-25 20:20 +0200 |
| Articles | 3 — 2 participants |
Back to article view | Back to linux.debian.user
Sometimes "time shift" after resuming from suspend Michael Lange <klappnase@freenet.de> - 2016-04-25 17:40 +0200
Re: Sometimes "time shift" after resuming from suspend Michael Biebl <biebl@debian.org> - 2016-04-25 19:10 +0200
Re: Sometimes "time shift" after resuming from suspend Michael Lange <klappnase@freenet.de> - 2016-04-25 20:20 +0200
| From | Michael Lange <klappnase@freenet.de> |
|---|---|
| Date | 2016-04-25 17:40 +0200 |
| Subject | Sometimes "time shift" after resuming from suspend |
| Message-ID | <rrQMa-8el-9@gated-at.bofh.it> |
Hi, here (Jessie with systemd) occasionally after the system is resumed from suspend mode the time is shifted forward by what appears to be exactly two hours. This does not occur always on resume, just sometimes, with no apparent reason. I have checked the syslog after this happened, but the only lines I found concerning the change of system time are these: Apr 25 02:19:45 localhost systemd[1383]: Time has been changed Apr 25 02:19:45 localhost systemd-sleep[2090]: System resumed. which are also there when the time was set correctly, so I guess they do not mean anything. Now I suspect that the two-hour shift is related to the hardware clock using local time instead of UTC (because of an existing WinXP Partition), which is here CEST (UTC + 2hrs.).The timezone is set to Europe/Berlin in /etc/timezone, in /etc/adjtime the respective line is set to LOCAL, which I believe should be ok. I also doubt that a general faulty configuration is responsible for the problem, since it happens only sometimes, in most cases the system time is handled correctly. Anyway, I have no idea what might cause this strange behavior, a web search did not show up something helpful either. The command I use to suspend the system is usually "systemctl hybrid-sleep", I don't think this makes a difference to a "normal" suspend, though I am not 100% sure. Does anyone have an idea what might be the cause for this? Best regards Michael .-.. .. ...- . .-.. --- -. --. .- -. -.. .--. .-. --- ... .--. . .-. Those who hate and fight must stop themselves -- otherwise it is not stopped. -- Spock, "Day of the Dove", stardate unknown
[toc] | [next] | [standalone]
| From | Michael Biebl <biebl@debian.org> |
|---|---|
| Date | 2016-04-25 19:10 +0200 |
| Message-ID | <rrSbg-172-1@gated-at.bofh.it> |
| In reply to | #167081 |
[Multipart message — attachments visible in raw view] — view raw
Am 25.04.2016 um 17:35 schrieb Michael Lange: > Hi, > > here (Jessie with systemd) occasionally after the system is resumed from > suspend mode the time is shifted forward by what appears to be exactly two > hours. This does not occur always on resume, just sometimes, with no > apparent reason. I have checked the syslog after this happened, but the > only lines I found concerning the change of system time are these: > > Apr 25 02:19:45 localhost systemd[1383]: Time has been changed > Apr 25 02:19:45 localhost systemd-sleep[2090]: System resumed. > > which are also there when the time was set correctly, so I guess they do > not mean anything. > > Now I suspect that the two-hour shift is related to the hardware clock > using local time instead of UTC (because of an existing WinXP Partition), Fwiw, with recent Windows editions it should be ok to enable UTC. There is a registry key which needs to be set. Dunno though, if XP is recent enough. > which is here CEST (UTC + 2hrs.).The timezone is set to Europe/Berlin > in /etc/timezone, in /etc/adjtime the respective line is set to LOCAL, > which I believe should be ok. I also doubt that a general faulty > configuration is responsible for the problem, since it happens only > sometimes, in most cases the system time is handled correctly. > > Anyway, I have no idea what might cause this strange behavior, a web > search did not show up something helpful either. > The command I use to suspend the system is usually > "systemctl hybrid-sleep", I don't think this makes a difference to a > "normal" suspend, though I am not 100% sure. > > Does anyone have an idea what might be the cause for this? For how long did the system sleep when this log message occurs? This probably just means, that your hwclock drifted enough so the clock needed adjustment. Michael -- Why is it that all of the instruments seeking intelligent life in the universe are pointed away from Earth?
[toc] | [prev] | [next] | [standalone]
| From | Michael Lange <klappnase@freenet.de> |
|---|---|
| Date | 2016-04-25 20:20 +0200 |
| Message-ID | <rrTh0-1RH-13@gated-at.bofh.it> |
| In reply to | #167087 |
On Mon, 25 Apr 2016 19:02:10 +0200 Michael Biebl <biebl@debian.org> wrote: > Fwiw, with recent Windows editions it should be ok to enable UTC. There > is a registry key which needs to be set. > Dunno though, if XP is recent enough. thanks, I didn't know that, seems like it's actually possible with XP, though there still seem to be a few pitfalls left (according to a quick web search). > For how long did the system sleep when this log message occurs? > This probably just means, that your hwclock drifted enough so the clock > needed adjustment. This doesn't seem to matter, the message is there even if I resume after some 20 seconds. Another observation I just made: I tested a simple script that will call ntpdate on resume to correct the broken change to the system time if necessary; for testing I called "date" to temporarily set the time one our backwards. When I the called systemctl hybrid-sleep it looked like the time was set *three* hours forward this time already while the suspend/hibernate was running, so it looks like systemd's suspend (or hibernate?) procedure is the evildoer here, not the resume. I figured that the additional hour in this try might be caused by the contents of /etc/adjtime that might have not been updated by my "date" call?! Anyway, I am still really puzzled how such an odd behavior is possible. Regards Michael .-.. .. ...- . .-.. --- -. --. .- -. -.. .--. .-. --- ... .--. . .-. I thought my people would grow tired of killing. But you were right, they see it is easier than trading. And it has its pleasures. I feel it myself. Like the hunt, but with richer rewards. -- Apella, "A Private Little War", stardate 4211.8
[toc] | [prev] | [standalone]
Back to top | Article view | linux.debian.user
csiph-web