Avamar: How to run and interpret the output of the Avamar CPS tool
Summary: This article shows how to run and interpret the output of the Avamar Checkpoint (CPS) show tool.
Instructions
About the tool
The 'cps' script is found on the Avamar Utility Node or single node grid in /usr/local/avamar/bin.
The script can be used to check the space consumption of Avamar data on a storage node. It does this by listing each checkpoint, and cur directory found in the /data0X partitions.
The tool reports the absolute and the relative sizes of the 'cur' and checkpoint directories on each node. It also shows the total space consumption of those directories within the file system.
-
Determine the space is consumed by '
cur' and by each checkpoint. -
Determine the data change occurred between checkpoints.
-
Identify which checkpoints are most beneficial to 'roll off' in order to recover from a high operating system capacity.
-
Compare the behavior of and sizes of data directories between Avamar nodes.
-
Check if any spurious checkpoint or cur directories exist.
How to run:
a. Log in to the Avamar Utility Node or Single node as admin.
On a single node grid, run the following command:
/usr/local/avamar/bin/cps
Example output:
GB used %use Total checkpoint usage by node:
1449.141 Total blocks on node Thu May 10 13:53:50 2018
1105.793 76.31 Total blocks available
279.817 19.31 cur Thu May 10 13:39:12 2018
58.221 4.02 cp.20180424154317 Tue Apr 24 15:43:32 2018
0.425 0.03 cp.20180424154252 Tue Apr 24 15:43:07 2018
0.423 0.03 cp.20180424154038 Tue Apr 24 15:40:53 2018
0.702 0.05 cp.20180424054447 Tue Apr 24 05:45:19 2018
0.703 0.05 cp.20180423230023 Mon Apr 23 23:34:52 2018
340.291 23.48 Total blocks used by dpn
On a multi node grid, load the admin keys per Avamar: How to Log in to an Avamar Server and Load Various Keys and run the following commands from /home/admin on the Avamar Utility Node:
cp /usr/local/avamar/bin/cps .
mapall copy cps
(This copies the script to all data nodes)
mapall ./cps
(This runs the script on all data nodes)
Example output:
(0.0) ssh -x admin@10.10.10.2 './cps'
GB used %use Total checkpoint usage by node:
1449.141 Total blocks on node Thu May 10 13:53:50 2018
1105.793 76.31 Total blocks available
279.817 19.31 cur Thu May 10 13:39:12 2018
58.221 4.02 cp.20180424154317 Tue Apr 24 15:43:32 2018
0.425 0.03 cp.20180424154252 Tue Apr 24 15:43:07 2018
0.423 0.03 cp.20180424154038 Tue Apr 24 15:40:53 2018
0.702 0.05 cp.20180424054447 Tue Apr 24 05:45:19 2018
0.703 0.05 cp.20180423230023 Mon Apr 23 23:34:52 2018
340.291 23.48 Total blocks used by dpn
(0.1) ssh -x admin@10.10.10.3 './cps'
GB used %use Total checkpoint usage by node:
1449.141 Total blocks on node Thu May 10 13:53:52 2018
1128.795 77.89 Total blocks available
271.074 18.71 cur Thu May 10 13:50:39 2018
45.136 3.11 cp.20120424154317 Tue Apr 24 15:43:30 2018
0.431 0.03 cp.20120424154252 Tue Apr 24 15:43:05 2018
0.430 0.03 cp.20120424154038 Tue Apr 24 15:40:51 2018
0.689 0.05 cp.20120424054447 Tue Apr 24 05:45:20 2018
0.696 0.05 cp.20120423230023 Mon Apr 23 23:34:52 2018
318.455 21.98 Total blocks used by dpn
(0.2) ssh -x admin@10.10.10.4 './cps'
GB used %use Total checkpoint usage by node:
1449.141 Total blocks on node Thu May 10 13:53:53 2018
1109.543 76.57 Total blocks available
278.917 19.25 cur Thu May 10 13:39:30 2018
56.916 3.93 cp.20120424154317 Tue Apr 24 15:43:30 2018
0.463 0.03 cp.20120424154252 Tue Apr 24 15:43:05 2018
0.461 0.03 cp.20120424154038 Tue Apr 24 15:40:53 2018
0.732 0.05 cp.20120424054447 Tue Apr 24 05:45:19 2018
0.730 0.05 cp.20120423230023 Mon Apr 23 23:34:52 2018
338.219 23.34 Total blocks used by dpn
-
-
-
The cur directory is consuming ~280 GB of space, equivalent to ~19% of the size of the data partitions.
-
Space usage is similar across all nodes (in the multinode environment)
-
Change rate between checkpoints is low.
-
-