Unsolved
This post is more than 5 years old
41 Posts
0
950
using AFTDs but still waiting for tapes
I'm using AFTD's (disk devices) for all backup jobs, but when the backup starts, Networker immediately shows us with this message "Waiting for (x) writable volume(s) to backup pool "xxxxx" tape(s) or disk(s) on backup server".
The drives are mounted in Linux and not full and also writing while networker is requesting for more tape(s) or disk(s). I also increased the sessions in the devices, but still the same issue.
Is this normal or something is wrong?
Backup server is Networker v7.6 on RedHat Linux 5
thanks in advance for any help, suggestions and or solutions.
DavidHampson-rY
294 Posts
0
October 20th, 2010 08:00
That's a standard message, after you get that does the backup start okay?
lpamintuan
41 Posts
0
October 20th, 2010 09:00
the backups starts okay, but somehow I noticed a slight backup throughput degradation and also the message is cuasing "Panic or Confusion" on my backup operators.
DavidHampson-rY
294 Posts
0
October 20th, 2010 09:00
When you talk about a "backup throughput degradation" how much is that degradation and what are you using as your baseline.
The message should no longer cause panic if they know it is an informational message - it is only an issue if it continues to repeat.
lpamintuan
41 Posts
0
October 21st, 2010 07:00
not sure also if this is causing this issue, some scheduled backups shows as started and running in NMC, but they are not writing to the disks, they are pending and might be "waiting for tapes" request is cleared although the disk device is writing and only with 5 savesets on it. The message stays untill all the backups were done.
ble1
14.3K Posts
1
October 25th, 2010 01:00
Could be that you parallelism settings (and few others) are limiting number of sessions... so as long as number of client sessions ready to run > what devices are configured to accept = you will see this message.
A typical "issue" is that you have AFTD devices labeled for several pools and associated against group in pool. In such case, NW will count number of available devices first and then spread them across without taking into account that not all devices might be used for that pool. If you use device restriction that you do not see that. I've been waiting for years to see that algorithm fixed, but haven't noticed any changed so far.