lubuntu-qa team mailing list archive
-
lubuntu-qa team
-
Mailing list archive
-
Message #03741
Re: zRAM broken on Raring?
Sorry, I overlooked that John is running Raring. I didn't think zRAM was
used in Raring, but of course, John can install and run it. I'm glad you
corrected that mistake by me.
Best regards
Nio
On 2013-09-28 00:08, Phill Whiteside wrote:
> hi Nio,
>
> he is running Raring. The bug we see in Saucy on 1227202 is totally
> un-related to the sudo parted -l issue.
>
> Getting the race issue sorted out on un-mounting the zram areas is what
> the bug fix is. With all the tests I've done, the
>
> Error: /dev/zram0: unrecognised disk label
>
> Error: /dev/zram1: unrecognised disk label
>
> Has remained until I used the 3.12rc kernel. We will go battle that
> issue on Monday to see if we can find the fix. Also do
> read https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1215379 as we
> may be barking up the wrong tree with zram, and should be using zswap.
>
> Joe cannot tell us which is better to use, just that zswap was in the
> 3.11.2 upstream kernel which has been imported into the latest
> 3.11.0-9.16 kernel.
>
> It was Unit193 who pointed that issue out, and whilst getting the race
> crash sorted out, it appears that zswap is the updated system.
>
> Regards,
>
> Phill.
>
>
> On 27 September 2013 22:54, Nio Wiklund <nio.wiklund@xxxxxxxxx
> <mailto:nio.wiklund@xxxxxxxxx>> wrote:
>
> On 2013-09-27 23:28, John Hupp wrote:
> > On Raring, output from 'sudo parted -l' includes:
> >
> > Error: /dev/zram0: unrecognised disk label
> >
> > And syslog shows a slew of errors:
> >
> > Lubuntu kernel: Buffer I/O error on device zram0, logical
> block 128247
> >
> > Syslog also indicates that half of memory was given to zram to
> form its
> > block device.
> >
> > Does this mean that half of memory is dedicated to something that
> isn't
> > working? And perhaps that machines will hang when swap is needed?
> >
> > I arrive at this line of questioning because I was testing an LTSP
> > client using a Lubuntu LTSP server configured with 1 GB, and when
> I drop
> > the client memory configuration to 256 MB, the *server* has hung on
> > several occasions when I was starting or stopping Firefox on the
> client
> > (though in one case this coincided with the startup of a SpiderOak
> > backup operation). I didn't think of the Magic SysRq keys at the
> time,
> > and nothing else was responding, so I did hard shutdowns.
> >
> > I saw a post from Phill Whiteside recently concerning a rush of
> activity
> > re a zram bug, but it seemed to be directed at Saucy.
> >
> > Are there solutions/workarounds?
> >
> You find a lot of details reading the comments about this bug.
>
> https://bugs.launchpad.net/bugs/1227202
>
> There is a kernel in the pipeline, that we think will solve most if not
> all of the problems. Until we get the kernel that can cooperate with
> zRAM, you can switch it off either manually or with crontab like this:
>
> guru@Lubuntu-Saucy-b2:~$ sudo crontab -l |tail -n3
> # m h dom mon dow command
> @reboot /sbin/swapoff /dev/zram*
> @reboot /sbin/rmmod zram
>
> Best regards
> Nio
>
> --
> Lubuntu-users mailing list
> Lubuntu-users@xxxxxxxxxxxxxxxx <mailto:Lubuntu-users@xxxxxxxxxxxxxxxx>
> Modify settings or unsubscribe at:
> https://lists.ubuntu.com/mailman/listinfo/lubuntu-users
>
>
>
>
> --
> https://wiki.ubuntu.com/phillw
References