Issue with DSSU Cleanup post relocation
Updated: 31 Mar 2012 | 18 comments
This issue has been solved. See solution.
Hi,
Master Server: Netbackup 6.5.4 Ent
Master Server Name: dakcmstnetbkp
Operating System: Solaris 10
NetBackup Version: 6.5.4
Cluster : No
We are having issues with disk staging cleanup post relocation to tape.
Cleanup does not execute inspite of reaching the high water mark resulting in the disk being full.
We checked the images on disk in catalog option and found it still shows as primary copy while the images on tape show as copy2.
Tried the following -
1) bpimage -cleanup -allclients
2) Tried creating new dssu for the same volume and fired a test backup to disk then executed manual relocation which duplicated the image to tape but did not expire the disk image
Sammeer
Discussion Filed Under:
Comments
The images stay on disk as
The images stay on disk as long as possible (or until they expire)
The disk images are copy 1 and are primary
The tape images are copy 2 and only become primary once the disk images are expired
Are you actually seeing failures as a result of the disk filling up?
http://www.symantec.com/docs/TECH66149
This doc covers disk cleanup behavoir in NetBackup - the key being the difference between the High and Low watermarks
However, if you are not duplicating your backups efficiently enough then NetBackup cannot do the cleanup and so will grow above the High Watermark, eventually filling the disk when backups will fail
The setting are just triggers and reply on verything else being in order to be of any effect
Hope this helps
Authorised Symantec Consultant
Don't forget to give a "Thumbs Up" or mark as "Solution" if someones advice has helped you.
Did this every work ? If it
Did this every work ?
If it did, what has changed ?
Martin
Also realised you use bpimage
Also realised you use bpimage -cleanup -allclients, but if they are not expired they will not cleanup from disk
If you want to cleanup expired disk images only use: nbdelete -allvolumes -force
But again, if they have not expired they won't go
Staging only clears down images when it has hit the high watermark and only then on the condition that the images have been duplicated to tape os make sure your duplications are all up to date and refer to the tech note in relation to High and Low Watermarks - what settings are you using and what size is the staging area?
Authorised Symantec Consultant
Don't forget to give a "Thumbs Up" or mark as "Solution" if someones advice has helped you.
The disk is full and already
The disk is full and already crossed the high water mark
Since this is a temporary staging area I do not see a reason for the images to remain on disk after relocation has successfully completed
No settings were changed as this has worked for last 2 yrs
Currently I am manually marking all relocated images to tape as primary and expiring the images in dssu
Sammeer S. Pednegar
Senior Solutions Specialist
Symantec Corporation
www.symantec.com
Mobile:
This is by design - the disk
This is by design - the disk staging area will remain full at all times (at least up to the High Water Mark)
This allows any restores needed to be done quickly from disk rather than from tape
If you expire everything it wil just fill up again
If you are running manual relocations and duplications are kicking in then your duplications are not up to date and you should look at your staging schedule
If you are not actually getting any Status 129 - Disk Storage Unit Is Full errors then there is nothing wrong with your setup
Authorised Symantec Consultant
Don't forget to give a "Thumbs Up" or mark as "Solution" if someones advice has helped you.
Is this a cluster, or, aare
Is this a cluster, or, aare there multiple interfaces on the master ?
OK, if you look in the expmgr log ( 272) (Debug / Diag at 6, do you see a High Water Mark event.
Something like this :
16/08/2011 22:29:40.160 V-272-6 [ExpMgrEventConsumer::notify] High Water Mark event received
vxlogview should be run with the options -d all -i 272
The logs involved for this part :
272 - expmgr (this is written into the 226 log)
226 - nbstserv
111 - nbemm
230 - rdsm
178 - dsm
mkdir /usr/openv/netbackup/logs/admin
Add VERBOSE = 5 to bp.conf
Without testing this, there couold be other logs involved, I think to start with .
1 Enable the 272 / 226/ 111 logs at debug/ diag 6
2 Create the 'fault' that is, allow the disk space to fill up to the HWM
3 Grab logs, run through vxlogview as shown
-i 111 -d all for the emm log
-i 226 -d -all for nbstserv log
+ admin log.
This is from memory, not able to test/ check at the moment, so apologies if I have made a mistake ...
Thanks,
Martin
I would like to see a bpdm
I would like to see a bpdm log as well...
NBU 6.5 DSSU cleanup behaviour is explained in this TN: http://www.symantec.com/docs/TECH66149
What kind of storage are you using for DSU?
Is it a dedicated lun / mount point for DSU only and not a repository for other kinds of data as well?
Supporting Storage Foundation and VCS on Unix and Windows as well as NetBackup on Unix and Windows.
Handy NBU links
Marianne is correct, the
Marianne is correct, the bpdbm log is involved.
In the interests of keeping things simple to start with, my intention was to look only to see if the HWM event is triggered, If this is NOT happeneing, then we know (approx) to look, if it is happening we would then move on to look at other areas, for example bpdbm.
It is difficult to know where to start - if we looked at every possible log in this 'process' - that is a backup that 'should' trigger a disk cleardown - there are probably 15-20 possible logs. Straight away, we can discount some of these, and are then left with xx possible logs, which is quite a lot in this instance ...
So, the question is, do we go for all of them, wth the possibility we only need some of them. This would ensure we get the correct logs, but, spend a lot of time and effort getting logs we don't in fact need.
Or ...
Do we go for some of them and hope we guess correctly ...
So, on reflection, let us get the bpdbm log in as well, as suggested by Marianne.
Martin
the DSSU is a dedicated lun /
the DSSU is a dedicated lun / mount point
Sammeer S. Pednegar
Senior Solutions Specialist
Symantec Corporation
www.symantec.com
Mobile:
Hi Marianne The understanding
Hi Marianne
The understanding about highwater mark and low water mark explained in the shared article is correct.
We have tried lowering the HWM to 10 from a previous value of 75 but the images refused to cleanup automatically.
Please find the bpbrmds, bpdm and vxlogmgr logs in the below mentioned fileshare link
https://fileshare.symantec.com
Folder - SymantecConnect
userid - Sagar_K
passwd - Sym@nt3c
Please Note-Logs have been compressed using 7z application which can be freely downloaded from http://www.7-zip.org/
Regards
Sammeer
Sammeer S. Pednegar
Senior Solutions Specialist
Symantec Corporation
www.symantec.com
Mobile:
Hi Sammeer, Due to a bit of
Hi Sammeer,
Due to a bit of confusion, we actually meant to ask also for the bpdm log - please accept out apologies for this.
We need this log, for the same time period as the other logs.
Martin
OK, I see you sent the bpdm
OK, I see you sent the bpdm log already - thx.
The HWM event is seen in the 226 log (74971 times in fact ... ) so I think we can say it works ...
I don't know which STU is not clearing down ??? but ...
Looking in bpdm ...
HighWater Mark 10, Low Water
HighWater Mark 10, Low Water Mark 5
This could strangle the system a little to be honest - I understand what you are trying but these settings are extreme and NetBackup may just not be able to do anything with them
Can I just clarify two things please...
1. Are all of your duplications kept up to date - what is the window on your disk staging schedule?
2. Could you give us the output of:
nbdevquery -listdv -D -stype BasicDisk
Just to double check everything
Authorised Symantec Consultant
Don't forget to give a "Thumbs Up" or mark as "Solution" if someones advice has helped you.
Hi Mark 1. Are all of your
Hi Mark
1. Are all of your duplications kept up to date - what is the window on your disk staging schedule?
- We have a daily schedule for relocation set to 10am
Please find the below output of nbdevquery -listdv -D -stype BasicDisk
Disk Volume Dump
name : <Internal_16>
id : <G:\netback9_Monthly_dssu>
diskpool : <netback9_Monthly_dssu::dakcmstnetbkp::BasicDisk>
disk_media_id : <@aaaam>
total_capacity : 2013384003584
total_phys_capacity : 2013384003584
free_space : 1056105869312
free_phys_space : 957278134272
potential_free_space: 0
committed_space : 0
precommitted_space : 0
nbu_state : 1
sts_state : 0
flags : 0xf
num_read_mounts : 0
max_read_mounts : 0
num_write_mounts : 1
max_write_mounts : 1
system_tag : <Generic disk volume>
Disk Volume Dump
name : <Internal_16>
id : </dss2>
diskpool : <Dakcmstnetbkp_dssu2::dakcmstnetbkp::BasicDisk>
disk_media_id : <@aaaaI>
total_capacity : 993989670912
total_phys_capacity : 993989670912
free_space : 856050805760
free_phys_space : 137938865152
potential_free_space: 0
committed_space : 0
precommitted_space : 0
nbu_state : 1
sts_state : 0
flags : 0xe
num_read_mounts : 0
max_read_mounts : 0
num_write_mounts : 1
max_write_mounts : 1
system_tag : <Generic disk volume>
Disk Volume Dump
name : <Internal_16>
id : <S:\>
diskpool : <netback9_full_dssu::dakcmstnetbkp::BasicDisk>
disk_media_id : <@aaaa3>
total_capacity : 982978129920
total_phys_capacity : 982978129920
free_space : 941469016064
free_phys_space : 41509113856
potential_free_space: 51280728064
committed_space : 0
precommitted_space : 0
nbu_state : 1
sts_state : 0
flags : 0xf
num_read_mounts : 0
max_read_mounts : 0
num_write_mounts : 1
max_write_mounts : 1
system_tag : <Generic disk volume>
Disk Volume Dump
name : <Internal_16>
id : </dss4>
diskpool : <Dakcmstnetbkp_dssu3::dakcmstnetbkp::BasicDisk>
disk_media_id : <@aaaaj>
total_capacity : 1932156461056
total_phys_capacity : 1932156461056
free_space : 1923511672832
free_phys_space : 8644788224
potential_free_space: 14720761856
committed_space : 0
precommitted_space : 0
nbu_state : 1
sts_state : 0
flags : 0xe
num_read_mounts : 0
max_read_mounts : 0
num_write_mounts : 1
max_write_mounts : 1
system_tag : <Generic disk volume>
Disk Volume Dump
name : <Internal_16>
id : <G:\netback9_Daily_dssu>
diskpool : <netback9_Daily_dssu::dakcmstnetbkp::BasicDisk>
disk_media_id : <@aaaan>
total_capacity : 2013384003584
total_phys_capacity : 2013384003584
free_space : 1056105869312
free_phys_space : 957278134272
potential_free_space: 0
committed_space : 0
precommitted_space : 0
nbu_state : 1
sts_state : 0
flags : 0xf
num_read_mounts : 0
max_read_mounts : 0
num_write_mounts : 1
max_write_mounts : 1
system_tag : <Generic disk volume>
Disk Volume Dump
name : <Internal_16>
id : </catalog/HSL_Daily>
diskpool : <Dakcmstnetbkp_dssu_HSL_Daily::dakcmstnetbkp::BasicDisk>
disk_media_id : <@aaaat>
total_capacity : 105432537088
total_phys_capacity : 105432537088
free_space : 105318130688
free_phys_space : 114406400
potential_free_space: 0
committed_space : 0
precommitted_space : 0
nbu_state : 1
sts_state : 0
flags : 0xe
num_read_mounts : 0
max_read_mounts : 0
num_write_mounts : 1
max_write_mounts : 1
system_tag : <Generic disk volume>
Disk Volume Dump
name : <Internal_16>
id : </catalog/HSL_Weekly>
diskpool : <Dakcmstnetbkp_dssu_HSL_Weekly::dakcmstnetbkp::BasicDisk>
disk_media_id : <@aaaas>
total_capacity : 105432537088
total_phys_capacity : 105432537088
free_space : 105318130688
free_phys_space : 114406400
potential_free_space: 0
committed_space : 0
precommitted_space : 0
nbu_state : 1
sts_state : 0
flags : 0xe
num_read_mounts : 0
max_read_mounts : 0
num_write_mounts : 1
max_write_mounts : 1
system_tag : <Generic disk volume>
Disk Volume Dump
name : <Internal_16>
id : <G:\netback4_full_dssu>
diskpool : <netback4_full_dssu::dakcmstnetbkp::BasicDisk>
disk_media_id : <@aaaaz>
total_capacity : 982970277888
total_phys_capacity : 982970277888
free_space : 940054462464
free_phys_space : 42915815424
potential_free_space: 18360924160
committed_space : 0
precommitted_space : 0
nbu_state : 1
sts_state : 0
flags : 0xe
num_read_mounts : 0
max_read_mounts : 0
num_write_mounts : 1
max_write_mounts : 1
system_tag : <Generic disk volume>
Disk Volume Dump
name : <Internal_16>
id : <S:\Weekly_Dssu\>
diskpool : <netback9_Weekly_dssu::dakcmstnetbkp::BasicDisk>
disk_media_id : <@aaaa6>
total_capacity : 982978129920
total_phys_capacity : 982978129920
free_space : 941469016064
free_phys_space : 41509113856
potential_free_space: 0
committed_space : 0
precommitted_space : 0
nbu_state : 1
sts_state : 0
flags : 0xf
num_read_mounts : 0
max_read_mounts : 0
num_write_mounts : 1
max_write_mounts : 1
system_tag : <Generic disk volume>
Disk Volume Dump
name : <Internal_16>
id : <S:\Monthly_Dssu\>
diskpool : <netback9_monthly_dssu::dakcmstnetbkp::BasicDisk>
disk_media_id : <@aaaa7>
total_capacity : 982978129920
total_phys_capacity : 982978129920
free_space : 941469016064
free_phys_space : 41509113856
potential_free_space: 4748288
committed_space : 0
precommitted_space : 0
nbu_state : 1
sts_state : 0
flags : 0xf
num_read_mounts : 0
max_read_mounts : 0
num_write_mounts : 1
max_write_mounts : 1
system_tag : <Generic disk volume>
Disk Volume Dump
name : <Internal_16>
id : </dss2/Weekly_Dssu/>
diskpool : <Dakcmstnetbkp_dssu2_Weekly::dakcmstnetbkp::BasicDisk>
disk_media_id : <@aaaa8>
total_capacity : 993989670912
total_phys_capacity : 993989670912
free_space : 856050805760
free_phys_space : 137938865152
potential_free_space: 0
committed_space : 0
precommitted_space : 0
nbu_state : 1
sts_state : 0
flags : 0xe
num_read_mounts : 0
max_read_mounts : 0
num_write_mounts : 1
max_write_mounts : 1
system_tag : <Generic disk volume>
Disk Volume Dump
name : <Internal_16>
id : </dss2/Monthly_Dssu/>
diskpool : <Dakcmstnetbkp_dssu2_Monthly::dakcmstnetbkp::BasicDisk>
disk_media_id : <@aaaa9>
total_capacity : 993989670912
total_phys_capacity : 993989670912
free_space : 856050805760
free_phys_space : 137938865152
potential_free_space: 0
committed_space : 0
precommitted_space : 0
nbu_state : 1
sts_state : 0
flags : 0xe
num_read_mounts : 0
max_read_mounts : 0
num_write_mounts : 1
max_write_mounts : 1
system_tag : <Generic disk volume>
Disk Volume Dump
name : <Internal_16>
id : </dss4/Weekly_Dssu/>
diskpool : <Dakcmstnetbkp_dssu3_Weekly::dakcmstnetbkp::BasicDisk>
disk_media_id : <@aaaaC>
total_capacity : 1932156461056
total_phys_capacity : 1932156461056
free_space : 1923511672832
free_phys_space : 8644788224
potential_free_space: 0
committed_space : 0
precommitted_space : 0
nbu_state : 1
sts_state : 0
flags : 0xe
num_read_mounts : 0
max_read_mounts : 0
num_write_mounts : 1
max_write_mounts : 1
system_tag : <Generic disk volume>
Disk Volume Dump
name : <Internal_16>
id : </dss4/Monthly_Dssu/>
diskpool : <Dakcmstnetbkp_dssu3_Monthly_::dakcmstnetbkp::BasicDisk>
disk_media_id : <@aaaaD>
total_capacity : 1932156461056
total_phys_capacity : 1932156461056
free_space : 1923511672832
free_phys_space : 8644788224
potential_free_space: 0
committed_space : 0
precommitted_space : 0
nbu_state : 1
sts_state : 0
flags : 0xe
num_read_mounts : 0
max_read_mounts : 0
num_write_mounts : 1
max_write_mounts : 1
system_tag : <Generic disk volume>
Disk Volume Dump
name : <Internal_16>
id : </dss4/impfiles/>
diskpool : <Dakcmstnetbkp_dssu3_impfiles::dakcmstnetbkp::BasicDisk>
disk_media_id : <@aaaaE>
total_capacity : 1932156461056
total_phys_capacity : 1932156461056
free_space : 1923511672832
free_phys_space : 8644788224
potential_free_space: 0
committed_space : 0
precommitted_space : 0
nbu_state : 1
sts_state : 0
flags : 0xe
num_read_mounts : 0
max_read_mounts : 0
num_write_mounts : 1
max_write_mounts : 1
system_tag : <Generic disk volume>
Disk Volume Dump
name : <Internal_16>
id : <G:\netback4_weekly\>
diskpool : <netback4_weekly_dssu::dakcmstnetbkp::BasicDisk>
disk_media_id : <@aaaaF>
total_capacity : 982970277888
total_phys_capacity : 982970277888
free_space : 940054462464
free_phys_space : 42915815424
potential_free_space: 0
committed_space : 0
precommitted_space : 0
nbu_state : 1
sts_state : 0
flags : 0xe
num_read_mounts : 0
max_read_mounts : 0
num_write_mounts : 1
max_write_mounts : 1
system_tag : <Generic disk volume>
Disk Volume Dump
name : <Internal_16>
id : <G:\netback4_monthly\>
diskpool : <netback4_monthly_dssu::dakcmstnetbkp::BasicDisk>
disk_media_id : <@aaaaG>
total_capacity : 982970277888
total_phys_capacity : 982970277888
free_space : 940054462464
free_phys_space : 42915815424
potential_free_space: 0
committed_space : 0
precommitted_space : 0
nbu_state : 1
sts_state : 0
flags : 0xe
num_read_mounts : 0
max_read_mounts : 0
num_write_mounts : 1
max_write_mounts : 1
system_tag : <Generic disk volume>
Sammeer S. Pednegar
Senior Solutions Specialist
Symantec Corporation
www.symantec.com
Mobile:
Could you please cover all of
Could you please cover all of the following points:
1. Could you clarify you staging window please - you say 10am each morning - what is the actual window and start time? Could you please provide the output of:
bpschedule -U
2. Which of the Disk Storage units do you have issues with?
I see that some have a figure for potential free space and some do not.
3. Please also provide the output of bpstulist -U
4. Do you actually experience failures?
Thanks
Authorised Symantec Consultant
Don't forget to give a "Thumbs Up" or mark as "Solution" if someones advice has helped you.
Hi Mark 1. Could you clarify
Hi Mark
1. Could you clarify you staging window please - you say 10am each morning - what is the actual window and start time? Could you please provide the output of:
bpschedule -U
- Output shared on Fileshare Userid - Sagar_K - Password - Sym@nt3c Folder name - SymantecConnect
2. Which of the Disk Storage units do you have issues with?
I see that some have a figure for potential free space and some do not.
- We are experiencing DSSU issue on all Disk Storage Units - even the ones that are connected to lan media servers.
3. Please also provide the output of bpstulist -U
- Output shared on Fileshare Userid - Sagar_K - Password - Sym@nt3c Folder name - SymantecConnect
4. Do you actually experience failures?
We do not experience any failures since we are performing manual cleanup of all images.
Sammeer S. Pednegar
Senior Solutions Specialist
Symantec Corporation
www.symantec.com
Mobile:
WOW! - so lets do these in
WOW! - so lets do these in importance order!!
Firstly these:
Dakcmstnetbkp_dssu2_Monthly
no window
Dakcmstnetbkp_dssu2_Weekly
no window
Dakcmstnetbkp_dssu3_Monthly_
no window
Dakcmstnetbkp_dssu3_Weekly
no window
netback9_Daily_dssu
no window
netback9_Monthly_dssu
no window
So none of these automatically duplicate any images - unless you run a manual relocation they do not go to tape so cannot clear images down automatically. If you do not manually relocate them and have exoired images to clear disk space then you have lost those backups.
Next this one:
Dakcmstnetbkp_dssu2
full window every 12 hours
This will kick in twice a day - so not too bad but see my reccomendations at the end
Next these:
Dakcmstnetbkp_dssu3
9am to 3pm every 10 hours
netback4_full_dssu
7am to 1pm every 12 hours
netback4_monthly_dssu
7am to 11am every 12 hours
netback4_weekly_dssu
7am to 11am every 12 hours
netback9_Weekly_dssu
7am to 11am every 12 hours
netback9_full_dssu
7am to 1pm every 12 hours
netback9_monthly_dssu
7am to 11am every 12 hours
Not great but as long as backups have all finished by 7am / 9am at least they run a duplication batch daily
Next these:
Dakcmstnetbkp_dssu_HSL_Weekly
8am to 10am every 7 days
Dakcmstnetbkp_dssu_HSL_Daily
8am to 10am every 7 days
These only run once a week but as you have every day specified they could run on any day so may not be able to clear anything down for a week.
So reccomendations ....
Set a window to the schedules for all those that do not have one
If you want the system to be almost as efficient as a SLP then set all DSSUs to have a full window (or at least a window from about an hour after your backups start in the evening to mid morning) with a frequency of 2 or 3 hours. If they do not have anything to do they will just run the parent job and finish with a blue man - far better than being behind on the duplications
I would reccomend changing ALL of you staging schedules immediately - I have the feeling based on what you have said that there is a possibility that you have lost data here.
Hope this helps
Authorised Symantec Consultant
Don't forget to give a "Thumbs Up" or mark as "Solution" if someones advice has helped you.
Disk Staging-Issue Resolved
We recycled the Netbackup services on the Master and now the images are cleaning up autoimatically.
The issue is resolved.
Sammeer S. Pednegar
Senior Solutions Specialist
Symantec Corporation
www.symantec.com
Mobile:
Would you like to reply?
Login or Register to post your comment.