OpenVZ Forum


Home » General » Support » incorrect size of root.hdd
incorrect size of root.hdd [message #51061] Tue, 14 January 2014 21:02 Go to next message
Grinvich is currently offline  Grinvich
Messages: 5
Registered: January 2014
Junior Member
We have CT with diskspace limit 100G.
volume of data inside of CT is 19G.
size of root.hdd is 100G.

then we do resize (decrease diskspace limit to 28G):

vzctl set 101 --diskspace 28G --save
dumpe2fs 1.41.12 (17-May-2010)
Changing balloon size old_size=75161927680 new_size=77309411328
Opening delta /vz/private/101/root.hdd/root.hdd
Successfully inflated balloon from 75161927680 to 77309411328 bytes
TRUNCATED: 1063 cluster-blocks (1114636288 bytes)
tune2fs 1.41.12 (17-May-2010)
Setting reserved blocks count to 366950
CT configuration saved to /etc/vz/conf/101.conf

now size of root.hdd is 80G.


then we do resize again (decrease diskspace limit to 25G):

# vzctl set 101 --diskspace 25G --save
dumpe2fs 1.41.12 (17-May-2010)
Changing balloon size old_size=77309411328 new_size=80530636800
Opening delta /vz/private/101/root.hdd/root.hdd
Successfully inflated balloon from 77309411328 to 80530636800 bytes
TRUNCATED: 2252 cluster-blocks (2361393152 bytes)
tune2fs 1.41.12 (17-May-2010)
Setting reserved blocks count to 327628
CT configuration saved to /etc/vz/conf/101.conf

now size of root.hdd is 77G


so, why size of root.hdd is more then diskspace limit and how we can fix size?
I remind that size of data inside of CT is 19G only!
Re: incorrect size of root.hdd [message #51064 is a reply to message #51061] Wed, 15 January 2014 18:45 Go to previous messageGo to next message
Paparaciz
Messages: 302
Registered: August 2009
Senior Member
Hi,
if the real data usage in CT is 19GB.
than you have to do:
vzctl compact ctid.

setting smaller --diskspace is completly wrong way...
and given all this I would sure make backup before compacting CT
Re: incorrect size of root.hdd [message #51068 is a reply to message #51064] Wed, 15 January 2014 19:21 Go to previous messageGo to next message
Grinvich is currently offline  Grinvich
Messages: 5
Registered: January 2014
Junior Member
thanks!

we have compacted 2 containers, but size of image decreased for 1 GB only and image of disk still have very incorrect size.

for example,

CT 101 :
79G root.hdd
df -h (inside of CT): 19G

CT 102 :
23G root.hdd
df -h (inside of CT): 5.1G
Re: incorrect size of root.hdd [message #51069 is a reply to message #51061] Wed, 15 January 2014 19:57 Go to previous messageGo to next message
Paparaciz
Messages: 302
Registered: August 2009
Senior Member
please report your kernel, vzctl, ploop versions
do you know who used all storage space?
maybe some service which deleted file but still using it?
Re: incorrect size of root.hdd [message #51070 is a reply to message #51069] Wed, 15 January 2014 20:21 Go to previous messageGo to next message
Grinvich is currently offline  Grinvich
Messages: 5
Registered: January 2014
Junior Member
2.6.32-042stab081.8

vzctl version 4.6

ploop-1.9-1.x86_64


we have restarted CT, so any services can't use deleted space
Re: incorrect size of root.hdd [message #51072 is a reply to message #51070] Thu, 16 January 2014 08:23 Go to previous messageGo to next message
Grinvich is currently offline  Grinvich
Messages: 5
Registered: January 2014
Junior Member
# ploop balloon discard -d /dev/ploop17875 --stat
Balloon size: 51200MB
Data size: 21047MB
Ploop size: 51200MB
Image size: 82846MB
Re: incorrect size of root.hdd [message #51074 is a reply to message #51061] Thu, 16 January 2014 19:16 Go to previous messageGo to next message
Paparaciz
Messages: 302
Registered: August 2009
Senior Member
I'm lazy to check it.

you didn't answered- do you know how you reached 100GB usage in CT?

maybe you did snapshot and didn't merged(deleted) it??
Re: incorrect size of root.hdd [message #51075 is a reply to message #51074] Thu, 16 January 2014 19:22 Go to previous message
Grinvich is currently offline  Grinvich
Messages: 5
Registered: January 2014
Junior Member
yes, there was snapshot, which not deleted from 27 December 2013.

and in this state CT was migrated to another node.

then was changed disksize limit from 100GB to 50GB (size of root.hdd was 100TB + about 15GB size of snapshot root.hdd{uuid} ).

and 2 days ago we deleted snapshot.
and after that, image file was decreased from 100GB to 80GB

Previous Topic: Killed screen/tmux sessions and terminated SSH
Next Topic: OpenVZ Kernel - RHEL 6.5 function
Goto Forum:
  


Current Time: Thu Aug 15 06:27:34 GMT 2024

Total time taken to generate the page: 0.02728 seconds