OpenVZ Forum


Home » Mailing lists » Users » Problems encountered increasing CT diskspace with layout=ploop
Problems encountered increasing CT diskspace with layout=ploop [message #45645] Mon, 26 March 2012 15:34 Go to next message
svensirk is currently offline  svensirk
Messages: 9
Registered: March 2012
Location: Hamburg
Junior Member
From: *parallels.com
Hi,

I have the same problem with a converted CT.
I converted from simfs to ploop using

vzctl convert <CTID> --layout ploop

After conversion I started the CT and tried to resize:

vzctl set <CTID> --diskspace 75G --save
Can't ioctl mount_point: No such file or directory
Failed to resize image: Can't ioctl mount_point: No such file or directory
[3]
CT configuration saved to /etc/vz/conf/<CTID>.conf

I can start/stop/chkpnt and restore the CT without any problems.
Only resize does not work.

I am Using
CentOS 6.2 64bit
vzctl version 3.1
ploop version 1.1
Kernel 2.6.32-042stab053.3

best regards,

Sirk

--


--
Re: Problems encountered increasing CT diskspace with layout=ploop [message #45652 is a reply to message #45645] Tue, 27 March 2012 11:11 Go to previous messageGo to next message
kir is currently offline  kir
Messages: 1645
Registered: August 2005
Location: Moscow, Russia
Senior Member

From: *parallels.com
> Hi,
>
> I have the same problem with a converted CT.
> I converted from simfs to ploop using
>
> vzctl convert <CTID> --layout ploop
>
> After conversion I started the CT and tried to resize:
>
> vzctl set <CTID> --diskspace 75G --save
> Can't ioctl mount_point: No such file or directory
> Failed to resize image: Can't ioctl mount_point: No such file or
> directory [3]
> CT configuration saved to /etc/vz/conf/<CTID>.conf

First of all, I would like to have such reports as bugs in
bugzilla.openvz.org rather than emails to users@. Bugzilla is
specifically designed to track bugs, while on a mailing list it could
easily be lost.

Now, can I ask you to debug this for me (since I am unable to reproduce it).

1. If you happen to have vzctl convert output, please show it. I am
especially interested in the line mentioning balloon.

2. Can I see the line in /proc/mounts regarding this mount? it can be
obtained by e.g.
cat /proc/mounts | grep -w <CTID> | grep ploop

Provided this, I will ask for more later.


Kir Kolyshkin
http://static.openvz.org/userbars/openvz-developer.png
Re: Problems encountered increasing CT diskspace with layout=ploop [message #45653 is a reply to message #45652] Tue, 27 March 2012 12:04 Go to previous messageGo to next message
svensirk is currently offline  svensirk
Messages: 9
Registered: March 2012
Location: Hamburg
Junior Member
From: *parallels.com
Hi,

Sorry for not posting in Buguzilla.
I found this thread an figured it was the right place to discuss.
Do you want me to open a bug in bugzilla ?

Here is the mount line:
/dev/ploop0p1 /srv/vz/root/102 ext4 rw,relatime,barrier=1,data=ordered 0 0


btw. resizing while the container is stopped does work.

Here is some output:
vzctl set 102 --diskspace 77G --save
Can't ioctl mount_point: No such file or directory
Failed to resize image: Can't ioctl mount_point: No such file or directory
[3]
CT configuration saved to /etc/vz/conf/102.conf
[root@testsystem1 ~]# cat /proc/mounts | grep -w 102 | grep ploop
/dev/ploop0p1 /srv/vz/root/102 ext4 rw,relatime,barrier=1,data=ordered 0 0
[root@testsystem1 ~]# vzctl stop 102
Stopping container ...
Container was stopped
Umounting fs at /srv/vz/root/102
Unmounting device /dev/ploop0
Container is unmounted
[root@testsystem1 ~]# vzctl set 102 --diskspace 77G --save
Adding delta dev=/dev/ploop0 img=/srv/vz/private/102/root.hdd/root.hdd (rw)
Mounting /dev/ploop0p1 at /srv/vz/private/102/root.hdd/root.hdd.mnt
fstype=ext4 data='balloon_ino=12,'
Growing dev=/dev/ploop0 size=150720512 sectors (new size=161480704)
Storing GPT
Executing: /sbin/resize2fs -p /dev/ploop0p1
resize2fs 1.41.12 (17-May-2010)
Das Dateisystem auf /dev/ploop0p1 ist auf
/srv/vz/private/102/root.hdd/root.hdd.mnt eingehängt;
Online-Grössenveränderung nötig
old desc_blocks = 5, new_desc_blocks = 5
Führe eine Online-Grössenänderung von /dev/ploop0p1 auf 20184827 (4k)
Blöcke durch.
Das Dateisystem auf /dev/ploop0p1 ist nun 20184827 Blöcke groß.

Executing: /sbin/tune2fs -r 1009254 /dev/ploop0p1
tune2fs 1.41.12 (17-May-2010)
Setze die Anzahl der reservierten Blöcke auf 1009254
Umounting fs at /srv/vz/private/102/root.hdd/root.hdd.mnt
Unmounting device /dev/ploop0
CT configuration saved to /etc/vz/conf/102.conf

After starting it again, it is not working anymore:

vzctl start 102
Starting container ...
Adding delta dev=/dev/ploop0 img=/srv/vz/private/102/root.hdd/root.hdd (rw)
Mounting /dev/ploop0p1 at /srv/vz/root/102 fstype=ext4
data='balloon_ino=12,'
Container is mounted
Adding IP address(es): 192.168.2.12
Setting CPU limit: 400
Setting CPU units: 1000
Setting CPUs: 4
Container start in progress...

[root@testsystem1 ~]# cat /proc/mounts | grep -w 102 | grep ploop
/dev/ploop0p1 /srv/vz/root/102 ext4 rw,relatime,barrier=1,data=ordered 0 0

[root@testsystem1 ~]# vzctl set 102 --diskspace 77G --save
Can't ioctl mount_point: No such file or directory
Failed to resize image: Can't ioctl mount_point: No such file or directory
[3]
CT configuration saved to /etc/vz/conf/102.conf

But if i do a chkpnt and restore it suddenly works:

[root@testsystem1 ~]# vzctl chkpnt 102
Setting up checkpoint...
suspend...
dump...
kill...
Checkpointing completed successfully
Umounting fs at /srv/vz/root/102
Unmounting device /dev/ploop0
Container is unmounted
[root@testsystem1 ~]# vzctl restore 102
Restoring container ...
Starting container ...
Adding delta dev=/dev/ploop0 img=/srv/vz/private/102/root.hdd/root.hdd (rw)
Mounting /dev/ploop0p1 at /srv/vz/root/102 fstype=ext4
data='balloon_ino=12,'
Container is mounted
undump...
Adding IP address(es): 192.168.2.12
Setting CPU limit: 400
Setting CPU units: 1000
Setting CPUs: 4
resume...
Container start in progress...
Restoring completed successfully
[root@testsystem1 ~]# cat /proc/mounts | grep -w 102 | grep ploop
/dev/ploop0p1 /srv/vz/root/102 ext4
rw,relatime,barrier=1,data=ordered,balloon_ino=12 0 0

[root@testsystem1 ~]# vzctl set 102 --diskspace 78G --save
Growing dev=/dev/ploop0 size=161480704 sectors (new size=163577856)
Storing GPT
Executing: /sbin/resize2fs -p /dev/ploop0p1
resize2fs 1.41.12 (17-May-2010)
Das Dateisystem auf /dev/ploop0p1 ist auf /srv/vz/root/102 eingehängt;
Online-Grössenveränderung nötig
old desc_blocks = 5, new_desc_blocks = 5
Führe eine Online-Grössenänderung von /dev/ploop0p1 auf 20446971 (4k)
Blöcke durch.
Das Dateisystem auf /dev/ploop0p1 ist nun 20446971 Blöcke groß.

Executing: /sbin/tune2fs -r 1022361 /dev/ploop0p1
tune2fs 1.41.12 (17-May-2010)
Setze die Anzahl der reservierten Blöcke auf 1022361
CT configuration saved to /etc/vz/conf/102.conf

It seems, that vzctl start does not really issue the balloon_ino option.

best regards and many thanks,

Sirk

2012/3/27 Kir Kolyshkin <kir@openvz.org>

> Hi,
>>
>> I have the same problem with a converted CT.
>> I converted from simfs to ploop using
>>
>> vzctl convert <CTID> --layout ploop
>>
>> After conversion I started the CT and tried to resize:
>>
>> vzctl set <CTID> --diskspace 75G --save
>> Can't ioctl mount_point: No such file or directory
>> Failed to resize image: Can't ioctl mount_point: No such file or
>> directory [3]
>> CT configuration saved to /etc/vz/conf/<CTID>.conf
>>
>
> First of all, I would like to have such reports as bugs in
> bugzilla.openvz.org rather than emails to users@. Bugzilla is
> specifically designed to track bugs, while on a mailing list it could
> easily be lost.
>
> Now, can I ask you to debug this for me (since I am unable to reproduce
> it).
>
> 1. If you happen to have vzctl convert output, please show it. I am
> especially interested in the line mentioning balloon.
>
> 2. Can I see the line in /proc/mounts regarding this mount? it can be
> obtained by e.g.
> cat /proc/mounts | grep -w <CTID> | grep ploop
>
> Provided this, I will ask for more later.
>



--
Satzmedia GmbH

Altonaer Poststraße 9
22767 Hamburg
Tel: +49 (0) 40 - 1 888 969 - 140
Fax: +49 (0) 40 - 1 888 969 - 200
E-Mail: s.johannsen@satzmedia.de
E-Business-Lösungen: http://www.satzmedia.de
Amtsgericht Hamburg, HRB 71729
Ust-IDNr. DE201979921
Geschäftsführer:
Dipl.-Kfm. Christian Satz
Dipl.-Inform. Markus Meyer-Westphal

--


--
Re: Problems encountered increasing CT diskspace with layout=ploop [message #45654 is a reply to message #45653] Tue, 27 March 2012 12:48 Go to previous messageGo to next message
kir is currently offline  kir
Messages: 1645
Registered: August 2005
Location: Moscow, Russia
Senior Member

From: *parallels.com
> Hi,
>
> Sorry for not posting in Buguzilla.
> I found this thread an figured it was the right place to discuss.
> Do you want me to open a bug in bugzilla ?

Yes please.
> Here is the mount line:
> /dev/ploop0p1 /srv/vz/root/102 ext4 rw,relatime,barrier=1,data=ordered 0 0

Right, no balloon_ino option...

>
>
> btw. resizing while the container is stopped does work.
>
> Here is some output:
> vzctl set 102 --diskspace 77G --save
> Can't ioctl mount_point: No such file or directory
> Failed to resize image: Can't ioctl mount_point: No such file or
> directory [3]
> CT configuration saved to /etc/vz/conf/102.conf
> [root@testsystem1 ~]# cat /proc/mounts | grep -w 102 | grep ploop
> /dev/ploop0p1 /srv/vz/root/102 ext4 rw,relatime,barrier=1,data=ordered 0 0
> [root@testsystem1 ~]# vzctl stop 102
> Stopping container ...
> Container was stopped
> Umounting fs at /srv/vz/root/102
> Unmounting device /dev/ploop0
> Container is unmounted
> [root@testsystem1 ~]# vzctl set 102 --diskspace 77G --save
> Adding delta dev=/dev/ploop0 img=/srv/vz/private/102/root.hdd/root.hdd
> (rw)
> Mounting /dev/ploop0p1 at /srv/vz/private/102/root.hdd/root.hdd.mnt
> fstype=ext4 data='balloon_ino=12,'
> Growing dev=/dev/ploop0 size=150720512 sectors (new size=161480704)
> Storing GPT
> Executing: /sbin/resize2fs -p /dev/ploop0p1
> resize2fs 1.41.12 (17-May-2010)
> Das Dateisystem auf /dev/ploop0p1 ist auf
> /srv/vz/private/102/root.hdd/root.hdd.mnt eingehängt;
> Online-Grössenveränderung nötig
> old desc_blocks = 5, new_desc_blocks = 5
> Führe eine Online-Grössenänderung von /dev/ploop0p1 auf 20184827 (4k)
> Blöcke durch.
> Das Dateisystem auf /dev/ploop0p1 ist nun 20184827 Blöcke groß.
>
> Executing: /sbin/tune2fs -r 1009254 /dev/ploop0p1
> tune2fs 1.41.12 (17-May-2010)
> Setze die Anzahl der reservierten Blöcke auf 1009254
> Umounting fs at /srv/vz/private/102/root.hdd/root.hdd.mnt
> Unmounting device /dev/ploop0
> CT configuration saved to /etc/vz/conf/102.conf
>
> After starting it again, it is not working anymore:
>
> vzctl start 102
> Starting container ...
> Adding delta dev=/dev/ploop0 img=/srv/vz/private/102/root.hdd/root.hdd
> (rw)
> Mounting /dev/ploop0p1 at /srv/vz/root/102 fstype=ext4
> data='balloon_ino=12,'

See -- balloon_ino option is passed on...

> Container is mounted
> Adding IP address(es): 192.168.2.12
> Setting CPU limit: 400
> Setting CPU units: 1000
> Setting CPUs: 4
> Container start in progress...
>
> [root@testsystem1 ~]# cat /proc/mounts | grep -w 102 | grep ploop
> /dev/ploop0p1 /srv/vz/root/102 ext4 rw,relatime,barrier=1,data=ordered 0 0

but it is not here. Hmm.

Can you check the dmesg please?

There should be something like
[597088.246208] ploop3: p1
[597088.288093] EXT4-fs (ploop3p1): mounted filesystem with ordered data
mode. Opts:
[597088.290410] EXT4-fs (ploop3p1): loaded balloon from 12 (0 blocks)
[597088.458139] CT: 123: started


Kir Kolyshkin
http://static.openvz.org/userbars/openvz-developer.png
Re: Problems encountered increasing CT diskspace with layout=ploop [message #45655 is a reply to message #45654] Tue, 27 March 2012 13:03 Go to previous message
svensirk is currently offline  svensirk
Messages: 9
Registered: March 2012
Location: Hamburg
Junior Member
From: *parallels.com
Hi,

here is the Bugzilla Bug ID: 2225
http://bugzilla.openvz.org/show_bug.cgi?id=2225

dmesg is:
[14476.673338] EXT4-fs (ploop0p1): mounted filesystem with ordered data
mode. Opts:
[14476.674206] EXT4-fs (ploop0p1): loaded balloon from 12 (0 blocks)
[14476.719799] CT: 102: started

Thanks for caring and all.
By the way. ploop is really awesome ! Exactly what I have been looking for
in OpenVZ for month.
It really solves a lot of problems with having containers on plain nfs.

many thanks,

Sirk

2012/3/27 Kir Kolyshkin <kir@openvz.org>

> Hi,
>>
>> Sorry for not posting in Buguzilla.
>> I found this thread an figured it was the right place to discuss.
>> Do you want me to open a bug in bugzilla ?
>>
>
> Yes please.
>
> Here is the mount line:
>> /dev/ploop0p1 /srv/vz/root/102 ext4 rw,relatime,barrier=1,data=**ordered
>> 0 0
>>
>
> Right, no balloon_ino option...
>
>
>
>>
>> btw. resizing while the container is stopped does work.
>>
>> Here is some output:
>> vzctl set 102 --diskspace 77G --save
>> Can't ioctl mount_point: No such file or directory
>> Failed to resize image: Can't ioctl mount_point: No such file or
>> directory [3]
>> CT configuration saved to /etc/vz/conf/102.conf
>> [root@testsystem1 ~]# cat /proc/mounts | grep -w 102 | grep ploop
>> /dev/ploop0p1 /srv/vz/root/102 ext4 rw,relatime,barrier=1,data=**ordered
>> 0 0
>> [root@testsystem1 ~]# vzctl stop 102
>> Stopping container ...
>> Container was stopped
>> Umounting fs at /srv/vz/root/102
>> Unmounting device /dev/ploop0
>> Container is unmounted
>> [root@testsystem1 ~]# vzctl set 102 --diskspace 77G --save
>> Adding delta dev=/dev/ploop0 img=/srv/vz/private/102/root.**hdd/root.hdd
>> (rw)
>> Mounting /dev/ploop0p1 at /srv/vz/private/102/root.hdd/**root.hdd.mnt
>> fstype=ext4 data='balloon_ino=12,'
>> Growing dev=/dev/ploop0 size=150720512 sectors (new size=161480704)
>> Storing GPT
>> Executing: /sbin/resize2fs -p /dev/ploop0p1
>> resize2fs 1.41.12 (17-May-2010)
>> Das Dateisystem auf /dev/ploop0p1 ist auf /srv/vz/private/102/root.hdd/**root.hdd.mnt
>> eingehängt; Online-Grössenveränderung nötig
>> old desc_blocks = 5, new_desc_blocks = 5
>> Führe eine Online-Grössenänderung von /dev/ploop0p1 auf 20184827 (4k)
>> Blöcke durch.
>> Das Dateisystem auf /dev/ploop0p1 ist nun 20184827 Blöcke groß.
>>
>> Executing: /sbin/tune2fs -r 1009254 /dev/ploop0p1
>> tune2fs 1.41.12 (17-May-2010)
>> Setze die Anzahl der reservierten Blöcke auf 1009254
>> Umounting fs at /srv/vz/private/102/root.hdd/**root.hdd.mnt
>> Unmounting device /dev/ploop0
>> CT configuration saved to /etc/vz/conf/102.conf
>>
>> After starting it again, it is not working anymore:
>>
>> vzctl start 102
>> Starting container ...
>> Adding delta dev=/dev/ploop0 img=/srv/vz/private/102/root.**hdd/root.hdd
>> (rw)
>> Mounting /dev/ploop0p1 at /srv/vz/root/102 fstype=ext4
>> data='balloon_ino=12,'
>>
>
> See -- balloon_ino option is passed on...
>
>
> Container is mounted
>> Adding IP address(es): 192.168.2.12
>> Setting CPU limit: 400
>> Setting CPU units: 1000
>> Setting CPUs: 4
>> Container start in progress...
>>
>> [root@testsystem1 ~]# cat /proc/mounts | grep -w 102 | grep ploop
>> /dev/ploop0p1 /srv/vz/root/102 ext4 rw,relatime,barrier=1,data=**ordered
>> 0 0
>>
>
> but it is not here. Hmm.
>
> Can you check the dmesg please?
>
> There should be something like
> [597088.246208] ploop3: p1
> [597088.288093] EXT4-fs (ploop3p1): mounted filesystem with ordered data
> mode. Opts:
> [597088.290410] EXT4-fs (ploop3p1): loaded balloon from 12 (0 blocks)
> [597088.458139] CT: 123: started
>
>
>


--
Satzmedia GmbH

Altonaer Poststraße 9
22767 Hamburg
Tel: +49 (0) 40 - 1 888 969 - 140
Fax: +49 (0) 40 - 1 888 969 - 200
E-Mail: s.johannsen@satzmedia.de
E-Business-Lösungen: http://www.satzmedia.de
Amtsgericht Hamburg, HRB 71729
Ust-IDNr. DE201979921
Geschäftsführer:
Dipl.-Kfm. Christian Satz
Dipl.-Inform. Markus Meyer-Westphal

--


--
Previous Topic: Re: [Announce] Kernel RHEL6 testing 042stab053.4
Next Topic: Re: [Announce] Kernel RHEL6 testing 042stab053.4
Goto Forum:
  


Current Time: Tue Nov 13 22:58:35 GMT 2018