Start a Conversation

Unsolved

This post is more than 5 years old

7493

December 3rd, 2015 15:00

Avamar Maintenance

Hello,

I have a question regarding Avamar maintenance window.  Currently the maintenance window is configured between 9am and 7pm.  Running the command STATUS.DPN I see the following:

Last checkpoint: took 33s (OK)

Last GC: took 2m 37s (OK)

Last hfscheck: finished after 12m 29s (OK)

I also see:

Maintenance windows scheduler capacity profile is active.

The maintenance window is currently running.

Currently running task(s): crunchwait

My question is how can I determine exactly how long maintenance is taking.  I'm interested in decreasing the 10 hour maintenance window.

Thanks in advance.

498 Posts

December 4th, 2015 13:00

Do you by chance have EMC's DPA?

you can get some good reports

I get this every day - the line will be RED if it fails.  And I can see how long things are taking.

Untitled.png

2 Posts

December 4th, 2015 16:00

Unfortunately, I do not have EMC DPA.  So I won't be able to leverage those reports. 

1.2K Posts

December 4th, 2015 17:00

I second using DPA for this.  It's terribly easy to get these reports emailed to you daily, or to view/summary weekly/monthly trends.

With DPA, you could look at the length of your GC and HFScheck over a long period of time, and feel reasonably safe about reducing the window.

Let us know if that helps!

Karl

2K Posts

December 7th, 2015 06:00

There is a support script called "sched.sh" that may do what you're looking for. The script prints a timeline of the last few days of maintenance, backups, replication, etc.. Here is some example output:

admin@testgrid:~/>: sched.sh

        12am                  1 1 1 1 1 1 1 1 1 1 2 2 2 2  EST

          0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3

2015/11/30 ............................................B.BB

2015/11/30 ............................................d.b.

2015/12/01 BBBBBBBBBBBBBBBBBBB.......b.bbb...b.........B.BB

2015/12/01 ..R............b..c.........................d.b.

2015/12/01 ..b...............g.............................

2015/12/01 ..................h.............................

2015/12/02 BBBBBBBBBBBBBBBBBBB.b.B...b.bbb.............B.BB

2015/12/02 ..R.............c.....b.....................d.b.

2015/12/02 ..bb............g...............................

2015/12/02 ................h...............................

2015/12/03 BBBBBBBBBBBBBBBBBBB.......b.bbb.............B.BB

2015/12/03 ..R.............b...........................d.b.

2015/12/03 ..b.............c...............................

2015/12/03 ................g...............................

2015/12/03 ................h...............................

2015/12/04 BBBBBBBBBBBBBBBBBBB.b.B...b...b.............B.BB

2015/12/04 ..R...................b.....................d.b.

2015/12/04 ..b.............................................

2015/12/05 BBBBBBBBBBBBBBBBBBB...........b.............B.BB

2015/12/05 ..R.............c...........................d.b.

2015/12/05 ..b.............g...............................

2015/12/05 ................h...............................

2015/12/06 BBBBBBBBBBBBBBBBBBB.........................B.BB

2015/12/06 ..R.............c...........................d.b.

2015/12/06 ..b.............g...............................

2015/12/06 ................h...............................

2015/12/07 BBBBBBBBBBBBBBBBBBB.............................

2015/12/07 ..R.............b...............................

2015/12/07 ..b.............c...............................

2015/12/07 ................g...............................

2015/12/07 ................h...............................

b=backup, h=hfs, c=CP, g=GC, r=Repl, d=ReplDest, e=restore

Uppercase/Reverse means failed.  v2.0

From the output, you can see that this test system is completing GC, checkpoints, and hfscheck within 30 minutes. You can also see that maintenance was missed on December 4 so if this were a production system, that would be something to look into.

You will likely need to pull the most recent version of the script down to your system because the one that ships with the software is somewhat out of date. The latest version of the script is available on the EMC FTP site.

5 Practitioner

 • 

274.2K Posts

March 22nd, 2016 07:00

Run avmaint sched crunchwait --ava

by default is 60 (1 hour)

498 Posts

March 23rd, 2016 09:00

I also vote for EMC DPA

I get the emails every day as well

And a Red line really stands out when a checkpoint fails

and it gave me a good idea of when and how long something runs

letting me schedules other activates better

there are also a number of other reports from Avamar I get.

like daily failures!

And it covers more then just Avamar.  if you have other EMC equipment you might very easily talk management in to setting up DPA for use by your IT department.

No Events found!

Top