Hello,

We are using softwareAG as Api management solution, we have launching a archive/purge job , but it takes more than 2 days and it still at “in progress state”, is there any possibility to stop it

Thank you.

any help pease

This one likely needs a ticket with support.

Does your purge job execute as an Integration Server service? If so, you can try cancelling or killing the thread associated with the service: Reverb

Hope this helps,
Percio

Hi Percio,

you can do that.

But if this is an archiving action from pub.monitor.archive folder, you will have to reset a value in the archiving database schema as otherwise the service detects that there is an archiving already running and stops without doing nothing.

Regards,
Holger

@Holger_von_Thomsen is there any archiving database, i guess that all data are saved on elsatic search database, but i dont know which index should i touch to stop the archive job.

Thank’s

@Percio_Castro1 im not sure, but i think that the archive job is not a service as it’s described in the official documentation, thank’s for, your help/

Archive job can be a service or stored procedure but the recommended and default behavior is stored procedure. There is an operation log table in archive db. You can monitor it to check if it is still running. You need to set OPERATION_STATUS to “COMPLETE” manually in order to stop it.

https://higherlogicdownload.s3.amazonaws.com/IMWUC/webMethods/Files3/9427307bc269bfbc33dea6dc50a2b12324295630_2_1035x679.png 1.5x, https://higherlogicdownload.s3.amazonaws.com/IMWUC/webMethods/Files/9427307bc269bfbc33dea6dc50a2b12324295630.png 2x" data-dominant-color="F2F2F3">

image1241×815 42.3 KB

You don’t really need to stop it unless it already stopped working. It rarely stops working, but when it happens its usually at first several runs. We are talking about Monitor Archive function right? Why do you want to stop it?

@ Engin SARLAK yes it’s abour monitor archive function, i had lauch 3 archive job, but always with the same result => all jobs are in in progress state even if i decided to archive juste one hour of a day

If it is the first time you are running the archive SP and if it has been bloated a lot, it will take a very long time to finish. It is expected and its nothing to be afraid of. Running the archive service again and again won’t change anything. All it is doing is to initiate archive SP, if it is running it won’t do anything. Check OPERATION_LOG table. If you see new entries, give it some time to finish. It can take longer then a week if you haven’t run it before. For better performance, you can implement a daily scheduled task for archive db service. If you run it daily, it doesn’t take that long, usually several minutes, that’s it.

Also if you want cancel it, run it with smaller batches. Bigger batches also work but it makes you feel like its not doing anything cause its taking very long time to finish big batches.

There may be some confusion here. API GW has an archive and purge function accessible via the console UI.

@eessaouira is this what you’re launching?

https://higherlogicdownload.s3.amazonaws.com/IMWUC/webMethods/Files3/ac61a52f7b98def010ffe8b9333cf9d62f9c749c_2_1033x639.png 1.5x, https://higherlogicdownload.s3.amazonaws.com/IMWUC/webMethods/Files/ac61a52f7b98def010ffe8b9333cf9d62f9c749c.png 2x" data-dominant-color="ECF2F7">

image1240×766 34.7 KB

Or this:

https://higherlogicdownload.s3.amazonaws.com/IMWUC/webMethods/Files3/cb88fae46d5539d207627fc82303ee13bcec5824_2_1035x432.png 1.5x, https://higherlogicdownload.s3.amazonaws.com/IMWUC/webMethods/Files/cb88fae46d5539d207627fc82303ee13bcec5824.png 2x" data-dominant-color="CBDDEA">

image1117×467 42.5 KB

You mentioned Elasticsearch and 1 hour of retention – which sounds like API GW archive/purge rather than Monitor.

@reamon the first one.

This is a pure guess, but make sure the file system has plenty of space. The Elasticsearch docs describe how archive/purge work and basically a big copy operation. We experienced similar “in progress”. After allocating more space it successfully completed, though still takes time.

@reamon i will allocate more space and wait for completion, thank you very much, is there any possibility to stop it or delete in progress jobs ?

Not that we found. They eventually dropped off the list.

Lots of ES magic behind the API GW facade. We’re still very much ES novices but each new challenge brings new learning opportunities.