Hi Bastien,
the problem persists with upowerd v0.99.0.
I'll go forth and file a bug report, then.
Regards,
vencik
On Thu, 2014-07-03 at 16:23 +0200, Bastien Nocera wrote:
> On Thu, 2014-07-03 at 15:12 +0200, Václav Krpec wrote:
> > Hi Bastien,
> >
> > please find strace & sysprof logs attach
On Thu, 2014-07-03 at 15:12 +0200, Václav Krpec wrote:
> Hi Bastien,
>
> please find strace & sysprof logs attached.
> The *.ok.* logs were taken after the problem disappears
> (AC adapter attached) for reference.
Looks like it could be the history writing indeed.
I'd advise you to upgrade to a
Yep, I forgot to take timestamps; sorry for that.
I've already told myself off for such a rookie mistake :-(
I've installed sysprof already, but will have to wait
another day till the battery gets drained again (the laptop
has some battery life).
Btw, I've cloned upower repo to check the code; wh
On Wed, 2014-07-02 at 14:40 +0200, Václav Krpec wrote:
> Hmmm, I wouldn't call 0.5 to 1 MB tiny amount...
It is when it's not synced to the disk and stays in the kernel's cache.
(By the way, you'll want to include "-ttt" in your strace flags, so that
time is printed).
Let me know when you've had
Oh, forgot to tell you the version: 0.9.23.
vencik
On Wed, 2014-07-02 at 12:51 +0200, Bastien Nocera wrote:
> On Wed, 2014-07-02 at 12:09 +0200, Václav Krpec wrote:
> > Hello everybody,
> >
> > as a follow-up of the previous post, this time I include
> > strace log.
> > upower seems to read bat
Hmmm, I wouldn't call 0.5 to 1 MB tiny amount...
vencik
On Wed, 2014-07-02 at 13:15 +0200, Bastien Nocera wrote:
> On Wed, 2014-07-02 at 13:11 +0200, Václav Krpec wrote:
> > Hi Bastien,
> >
> > you are right; the polling looks the same as now (I started to charge,
> > so the problem disappears)
On Wed, 2014-07-02 at 13:11 +0200, Václav Krpec wrote:
> Hi Bastien,
>
> you are right; the polling looks the same as now (I started to charge,
> so the problem disappears).
> However, I notice something else that is quite different:
> the amount of writing to /var/lib/upower/history- files.
> And
Hi Bastien,
you are right; the polling looks the same as now (I started to charge,
so the problem disappears).
However, I notice something else that is quite different:
the amount of writing to /var/lib/upower/history- files.
And quite substantial amounts of data, judging from the allocations.
Is
On Wed, 2014-07-02 at 12:09 +0200, Václav Krpec wrote:
> Hello everybody,
>
> as a follow-up of the previous post, this time I include
> strace log.
> upower seems to read battery info in busy loop:
> it appears to me that there's a failure in an attempt to recvmsg (EAGAIN)
> (that might be the bu
