Unsolved
This post is more than 5 years old
16 Posts
0
843
November 26th, 2012 03:00
save -c possible memory leak?
Hi there,
On one of our rhel 5.7 servers networker process "save -c" starts with a few MBs of memory and after 48 hours its memory usage jumps to 28GB. Server then crashes. I searched powerlink but cannot find anything. What causes this? Can we limit memory usage used by networker processes?
Networker version: 7.6.3
Linux client: RHEL 5.7 x86_64
0 events found
No Events found!


ble1
6 Operator
•
14.4K Posts
•
56.2K Points
0
November 26th, 2012 04:00
Try 7.6.4.2 client (or higher). I use that one with Linux clients (5.7 included) and have no issues (on the other hand, I don't have any backup running for so long).
robotics223
16 Posts
0
November 26th, 2012 06:00
We have dozens linux clients without problem, but this one have a few million files for backup, that's why it takes too long to backup. Can we upgrade clients to 7.6.4.x without upgrading server?
ble1
6 Operator
•
14.4K Posts
•
56.2K Points
0
November 26th, 2012 14:00
Yes, you can... however, if big number of files would be the issue, most likely I would go for different backup method (eg. SnapImage or storage snapshots).
robotics223
16 Posts
0
November 28th, 2012 22:00
I upgraded the client and still same issue, any advice?
Bebo2k
544 Posts
0
November 28th, 2012 23:00
Hi,
I have some question, You are using client side or server side backups ?
What is the backup command for configured in this client resource ?
Is this client with multiple network interfaces or multiple hostnames ?
You mentioned that this machine hold million of files , did you used to backup the machine full backup recently ? meaning when was the last successfull full and incremental backups or this machine has never been backed up ?
Waiting your kind replies.
Ahmed Bahaa
robotics223
16 Posts
0
November 29th, 2012 03:00
This one is using server side backups (backup is initiated from server).
Configured backup comman is save -c client_name
İt has both multiple interfaces with bonding and multiple hostname.
Last successful full backup is two months old, we continued incremental for two months and now full is failing.