I have recently upgraded to SnapCreator 4.0.0 and now the Job Monitor stops showing new jobs after the size limit (the default 100 in my case) is reached.
That means I have now exactly 100 finished jobs in my Job Monitor, which won't go away and don't make space for new jobs. I would expect when new jobs run that the Job Monitor drops any jobs older than the 100th to show new jobs, but this doesn't happen.
So since yesterday I cannot see any new backup jobs in the Job Monitor (without enlarging, but this would eventually lead to the same problem again), I can only see the current status by checking logs periodically.
Is this intended behavior and if so, can I delete the job monitor entries manually somehow? I didn't find any meaningful documentation the Job Monitor.
We are going to look into this and try and reproduce the issue. Can you provide exact steps you did from upgrade to the issue?
Right now we dont have a workaround we need to reproduce and investigate. I would say for now (if possible) revert back to SC 3.6 until we have this issue understood and can provide a resolution (unless you are OK with running it like this till we have resolution).
If this is an urgent issue then please open an NGS case, this places the appropriate level of urgency on the situation.
On behalf of the team we are really sorry you are having this issue and will get to bottom of it.
I followed the exact steps described for upgrading from 3.6 in the 4.0 Admin Guide. I only noticed the issue after the 100 jobs have been reached in the Monitor.
I yesterday even took the time to reinstall SC 4.0 from scratch (under a RedHat 6.3 instance) and re-create the profiles and jobs manually. However, the behaviour of not showing any new jobs after the size threshold has been reached in the Job Monitor stays the same
It's not a deal breaker currently, because scheduled jobs are still running and DFM event generation works so this will notify us correctly about failed backups.
But I'm looking forward to a solution, because it's still a weird behaviour which shouldn't occur I think.
I tried to reproduce this issue and I saw similar behavior, Job monitor stopped showing new jobs after the limit was reached.
But, I waiting for sometime and the jobs started showing up as expected and after that I could see the latest jobs.
So the point is once the job monitor size is reached the new jobs don't show up for sometime but then after a while they start showing up and then everything works fine.
I also saw this behavior after SC server re-start.
We are looking more deeper into the root cause of this behavior but I wanted to let you know that if you keep your jobs running for a while they will eventually start showing up.
Let us know if you observe the same in your environment or not.
We will keep you updated.
unfortunately my Job Monitor didn't resume to work... The last entry is still from last week (13 June) and no new entries appeared since then. The jobs themselves all ran fine until now, and event generation in DFM works without problems.
Speaking of DFM event generation: Is there a way to let SnapCreator not create "Critical" events for failed jobs, but only "Error" events?
Okay, so I guess that was just the behavior in my machine.
Just to get an idea of your scenario/environment, how many configuration do yu have and how are they scheduled?
Unfortunately there is no way right now that you can change it from "Critical" to "Error". we are making a note of it and we will record it as a feature request from you.
Currently we have 3 Profiles with 15 configurations in total. Each configuration is currently scheduled once a day for snapshot creation. All are connecting to SC agents in Windows servers, where a few use the smsql plugin, some only a 'sdcli snap create' command for application drives, vibe also for a few vSphere Datastores and two trigger OSSV transfers. All of the jobs (except the two OSSVs, obviously) additionally trigger a SnapVault update.
This will expand over the coming weeks, as we are currently in the progress of replacing all the local server schedules and home-grown scripts with SnapCreator, as it seems a very comprehensive solution to me for our heterogenous environment and I try to convince other clients' environments of the migration, too
This is also the reason for the question about the "Critical" event. We have Alert generation for Critical events enabled to send these to a SMS gateway so the on-call person receives a message about these. But we don't want to have them receiving text messages for every failed backup.
I have discovered another issue: When running a job in the scheduler manually (by clicking on Run), if the Start Date is in the past (which it is always when you defined the schedules not at the same day) the job is not starting because "The Start Date is not valid".
Running them automatically at the scheduled time however seems to work fine.