This post is more than 5 years old
27 Posts
1
15288
How to "refresh" a snapshot using CLI?
I need to create a set of commands for our AIX Sysadmins that will allow them to run scripts against an XIO array to refresh a snapshot on demand. They do not want me to re-map/mask the snapshot devices every time. They are fine with stopping the DB and unmounting the filesystems. They did just fine with TimeFinder on Vmax so I have no trust issues with giving them an account on the XIO. And before anyone suggests it, we are having issues getting Appsync running so I'm steering away from that. Since they're unix guys they want a command, not a GUI solution.
I am not a fan of RESTful API. I have plenty of other things to do and becoming a JSON programmer is not one of them. The documentation for RESTFUL is awful. The doc for XIO commands is awful. There are no cookbooks. There are no models.
I'm told I can "refresh" an existing snap but I can't find any XMS CLI (or RESTFUL API) commands to "refresh". Is this a single command? Is it an operand of a command? Is it a series of commands? I'm lost. Any help would be appreciated. Thanks.
mhageness
1 Message
0
December 23rd, 2015 07:00
Avi,
I've been following this thread and I was curious if you will be posting when the Patch is available to EMC customers? Any word on a release date as well?
Thanks!
Kumar_A
727 Posts
0
December 23rd, 2015 18:00
We are in the final stages – still waiting for an exact date though. In general, your XtremIO Sales/SE should have the most up to date information on this. We do communicate these details with them.
CFrank2
10 Posts
0
January 4th, 2016 07:00
I see that 4.0.2-65 was released on 12/27/15, but while browsing through the release notes I didn't notice any changes for REST APIs. However, I did see a new 4.0.2 REST API guide published, but didn't notice any difference in snapshotting behavior. Do you have any info on this?
Thanks.
CFrank2
10 Posts
0
January 5th, 2016 06:00
I installed 4.0.2-65 on our lab XtremIO and have been trying to get this functionality to work, but I'm not getting anywhere. It's behavior hasn't changed from 4.0.1. Unless I'm doing something wrong, I don't think this is it. Here's an example of my code:
Curl -X POST -d '{"cluster-id":1,"from-consistency-group-id":2,"to-snapshot-set-id":12,"no-backup"}' -u username:password -k -s https://XMS-IP/api/json/v2/types/snapshots
The command still creates a new snapshot-set and new snaps, leaving the old ones there. I've tried adding the "no-backup" option but then I get an error code 400, "Invalid request."
Kumar_A
727 Posts
0
January 5th, 2016 18:00
You need to provide a value for “no-backup”.
As the Restful API guide says, if ‘no-backup’ is included in the request, the source object is removed (regardless of the value entered for this property).
Kumar_A
727 Posts
0
January 5th, 2016 18:00
The RestAPI is fully backward compatible and the snapshot functionality remains the same in 4.0.2. There were some new parameters included in the RestAPI calls with XtremIO 4.0.2 release and hence the new version of the guide.
TimGoodman
2 Posts
0
January 12th, 2016 14:00
Marvin,
How come you step through the volumes one at a time to delete them instead of just deleting the ${SSOLD} snapshot set using the remove-snapshot-set command?
Thanks
marvin_keys
27 Posts
0
January 12th, 2016 18:00
The commands were provided by Scott H. from EMC. I am not that skilled in XIO CLI yet to have re-engineered them for improved efficiency.
In reference to another post regarding the "no-backup" option, IIRC, this particular option actually caused the problem. It is supposed to be fixed in the latest fix that is now available. Caution: if you are an EPIC shop do not apply the patch until you talk to Steve Lazer of EMC (EPIC people will know him). He is EMC's EPIC specialist and recommends you do NOT install this patch. We are fortunate enough to have (2) XIO clusters, one EPIC, one non-EPIC. We will be applying the full patch to the non-EPIC cluster to test it.
dynamox
1 Rookie
1 Rookie
•
20.4K Posts
0
January 13th, 2016 22:00
i am curious what's in the patch that is apparently not appropriate for EPIC shops ?
marvin_keys
27 Posts
0
January 14th, 2016 06:00
Perhaps another someone in a long line of Diva's has yet to put their blessing on it?
dynamox
1 Rookie
1 Rookie
•
20.4K Posts
0
January 14th, 2016 12:00
Caché specialist
AranH1
2.2K Posts
0
January 19th, 2016 13:00
I don't believe so. Cerner uses Oracle as their backend database which is a more common and understood workload on XtremIO compared to InterSystems Cache.
dynamox
1 Rookie
1 Rookie
•
20.4K Posts
0
January 19th, 2016 13:00
Hi Aran,
long time no talk indeed, how's them goats
Do you know if the same restrictions applies to Cerner ?
AranH1
2.2K Posts
0
January 19th, 2016 13:00
Dynamox,
(long time no talk!)
Whether the array is VNX, VMAX, or XTremIO, the EMC Epic Healthcare team doesn't approve a code level for deployment in Epic shops until internal testing has been completed under Epic workloads.
echolaughmk
522 Posts
0
January 19th, 2016 18:00
Does an XtremIO Uptime Bulletin exist for XtremIO yet like EMC has for the other products that details the relative percentage of customer adopting the different code levels so customers can compare? This would be very helpful to have in understanding what the latest target code is and the interoperability testing with other products.
If it does exist and I just couldn't find it on the support site, feel free to just shoot me the link. I posted a seperate discussion as well, but figured it played here with the recent replies. Thanks.