Closed GoogleCodeExporter closed 9 years ago
i resolve , with change the userid in /etc/passwd for hacluster
hacluster:x:0:0:ha_cluster user:/var/lib/pacemaker/cores/hacluster:/sbin/nologin
but this is not a good solution, can you help me?
Original comment by dotcom...@gmail.com
on 4 Apr 2014 at 4:18
Hi,
I just looked at a machine running an older version of ESOS and it definitely
appears to be a permissions problem. On the older box I see files owned by
'hacluster' in /dev/shm and not on a new build (as you described). I'm thinking
right now it could be related to the permissions on the /dev/shm mount; I think
these were updated not too many revisions ago. I'll look into it this evening.
In the mean time, I'm updating Pacemaker to 1.1.11... its something I've been
meaning to do for a while, and this reminded me. =)
So, I'll build an image on my test machine with the new version and see what
needs to be fixed. Should have something for ya before the weekend is over.
--Marc
Original comment by msmith...@gmail.com
on 4 Apr 2014 at 7:44
I've fixed this in r626:
https://googledrive.com/host/0B-MvNl-PpBFPbXplMmhwaElid0U/esos-0.1-r626.tar.xz
Could you please try the new version and confirm?
Thanks,
Marc
Original comment by msmith...@gmail.com
on 5 Apr 2014 at 5:37
ok, fixed permission.
but now i have:
local2.log:Apr 7 22:16:35 esosc2 corosync[2129]: [QB ] server name:
votequorum
local2.log:Apr 7 22:16:35 esosc2 corosync[2129]: [QB ] server name: quorum
local2.log:Apr 7 22:50:11 esosc2 corosync[2129]: [QB ] couldn't create
file for mmap
local2.log:Apr 7 22:50:11 esosc2 corosync[2129]: [QB ]
qb_rb_open:quorum-response-2129-2256-29: No space left on device (28)
local2.log:Apr 7 22:50:11 esosc2 corosync[2129]: [QB ] shm connection
FAILED: No space left on device (28)
local2.log:Apr 7 22:50:11 esosc2 corosync[2129]: [QB ] Error in connection
setup (2129-2256-29): No space left on device (28)
messages:Apr 7 22:16:35 esosc2 corosync[2129]: [QB ] server name: cmap
messages:Apr 7 22:16:35 esosc2 corosync[2129]: [QB ] server name: cfg
Original comment by i...@ucspat.org
on 7 Apr 2014 at 8:51
i just change fstab but not work!!
tmpfs /dev/shm tmpfs size=128m,mode=1777 0 0
Original comment by i...@ucspat.org
on 7 Apr 2014 at 10:31
ok, if i remount with size=128M working fine!
but on every reboot this is lost.
i don't know why, i just write this size in fstab and save it.
Original comment by dotcom...@gmail.com
on 8 Apr 2014 at 10:33
I'll make the change in trunk... for the time being, you have to do this to
make the change stick:
mount /mnt/root
vi /mnt/root/etc/fstab
umount /mnt/root
Its a chicken-and-egg problem... the core file systems are mounted before the
USB configuration is sync'd.
--Marc
Original comment by msmith...@gmail.com
on 9 Apr 2014 at 1:35
I committed this and the new image should be posted in about an hour.
--Marc
Original comment by msmith...@gmail.com
on 11 Apr 2014 at 1:11
Original issue reported on code.google.com by
dotcom...@gmail.com
on 4 Apr 2014 at 4:08