The Recoverpoint GUI has improved a lot over time, and the current incarnation in the “Unisphere” framework is the best yet, by a long way. Some operations are still a bit painful, and I’ve found myself this week cooking up some RP-CLI + Powershell goodness to get details in a fast, concise manner.
The task was simple. A new set of Clusters had been built, copious amounts of Consistency Groups created and the first series of transfers (full inits) were on the way over a relatively small link.
Everything that could be done to squeeze every last drop of performance out of the link was done. Compression, dedup etc, and I wanted to monitor the progress of the inits/transfers and RPAs.
That’s easy done through the GUI, but I wanted a summary of the 6 CG’s I had launched. Tedious in the GUI to flip between them, and it wasn’t going to cut it.
This is easily accomplished using RP CLI in a Powershell session, running commands to the clusters via plink. In the Powershell session allows you to leverage all the goodness therein (expressions, arrays, etc. All the thing RP ClI is missing! ).
This first example will retrieve the CG Statistics via the get_group_statistics cmd;
|
1 |
./plink.exe -ssh PRODRP -P 22 -l admin -pw pass get_group_statistics group=CGNAME |
If you omit the ‘group=‘ option, you’ll get all the CG’s stats returned.
Next, we’ll get the CG State, via the get_group_state cmd; (again, if you omit the ‘group’ you’ll get them all)
|
1 |
./plink.exe -ssh PRODRP -P 22 -l admin -pw pass get_group_state group=CGNAME |
Since my main need was to monitor the transfer state, the next cmd dumps the entire CG stat data to a file, reads it back via ‘Get-Content‘ then ‘greps’ the input via ‘Select-String‘ for the INITIALIZING string and returns a 2 line context from the result, above and below.
|
1 |
./plink.exe -ssh PRODRP -P 22 -l admin -pw pass get_group_state > c:\emcfiles\gs.txt ;gc c:\emcfiles\gs.txt |select-string -Pattern 'INITIALIZING' -Context 2 |
The last thing I was keeping an eye on was the state of the RPA’s. Just to be sure they remin healthy under peak loads. get_rpa_statistics was my friend here. Omit the ‘cluster’ and/or ‘rpa’ options to return all of them
|
1 |
./plink.exe -ssh PRODRP -P 22 -l admin -pw pass get_rpa_statistics cluster=PRODRP rpa=rpa4 |
there’s a plethora of stats, info and data there for the taking, in a manner that I find far easier to work with.
Job done !
As soon as I noticed this website I went on reddit to share some of the love with them.
Great post. I used to be checking continuously this blog and I am inspired! Very helpful info particularly the final section 🙂 I handle such information much. I used to be looking for this particular info for a very long time. Thank you and good luck.
Excellent article! We are linking to this particularly great post on our site.
Keep up the good writing.
Can all this be automated into one script so it can email out
Probably can, I haven’t gotten around to doing it though.
I like a challenge, so might just have a go at it soon 🙂