When looking at the list of backup jobs, it would be useful if one could click some column header to have the jobs sorted/listed in the order which they would run in. Purely for the purpose of getting an overview.
For example if I have jobs A, B, C, and D and they are set to execute after each other in the order A, C, D, B, that's how I want to be able to list them.
As it is now one have to follow/track the "Name" and "Next run" columns which can be confusing with a bunch of jobs
I’ve already created a script that outputs in chronological order the name of the jobs that will be executed during the current day and also the time when these jobs will be run. If you modify it a little bit, I believe, you will be able to achieve your goals.
v.Eremin, I'm not sure why you merged my post with this thread, from what I can tell the initial post here is completely different from what I suggested. I am not talking about the historical timeline, I am only talking about the list of backup jobs in the "Backup & Replication" view, under Jobs -> Backup in the "menu tree" to the left. Powershell isn't relevant either, sorry
Please "unmerge" it if possible. Or shall I make a new thread?
Actually, it was me and I still think that this feature request corresponds with the one discussed in the referred topic (listing jobs according to their chained runs can be a part of the mentioned historical view).
However, we can keep it separate if you wish. Thanks for the feedback!
Just to clarify, to avoid any misunderstandings, here's a screenshot of where I am asking for this feature to be added:
I honestly don't care about the/a history view, since in the history view you can already simply sort on the "Start time" column to see the order in which the jobs were run. But you can't do that here in the backup jobs "definition" list, and that's why I'm asking for it.
Being able to sort on the expected "start order" would make it easier to locate the jobs you want to edit or look at in this list (especially in bigger installations than this example one).
rawtaz wrote:Being able to sort on the expected "start order" would make it easier to locate the jobs you want to edit or look at in this list (especially in bigger installations than this example one).
Yep, I fully understand what you are talking about. Makes sense.
Regarding chained jobs in general, I'm using a different approach at all with very smaill and simple powershell script:
Community Forums • View topic - Veeam Backup Script to run multiple jobs in sequence http://forums.veeam.com/viewtopic.php?f ... har#p63202
Using such script has some benefits:
I can get a daily single summary report of all job runs instead.
There is no dependency between jobs.
I manually start a single job from the GUI (for example b4 maintenance of specific VM), without running all other chained job.
I can add/remove jobs to the script without thinking or modifying other jobs.
However - I would still look for a feature in the product itself which might have in the future a "job group" option, which would allow us to better schedule and monitor jobs in sequence.
yizhar wrote:However - I would still look for a feature in the product itself which might have in the future a "job group" option, which would allow us to better schedule and monitor jobs in sequence.
The product already technically allows this, well, at least partially. If you configure the proxy and/or repository to limit the number of parallel processes to 1 you can effectively start all of the jobs with 1 minute intervals and they will run in sequence. Perhaps it's not exactly what your looking for, but it's generally much easier than attempting to leverage linked jobs. I'm curious if you could describe the reasons for running jobs in sequence. Why not just put all of the VMs that need to be run in sequence in a single job since they can be ordered in the job? Is there some reason this doesn't work for you?
Reasons for running several smaller jobs vs single big one:
* I prefer to have the backups in several separated folders vs one big VBK file.
I know that I get dedup per job only and it is fine with me because of other benefits of separated jobs.
* I'm dividing jobs to server types, such as:
SQL
DC
and also per customer (we run a mini cloud farm for few clients).
This allows me to start a specific job as needed out of the regular schedule, for example before maintenance of specific servers.
It also gives me more control over some options such as guest processing, restore point to keep, and even destination repository if I have several to play with.
I have the freedom to run a specific job on demand as needed, and also the option to run them all.
In regards to running jobs in sequence vs parallel processing:
Well I have a long backup window, and the backups are running quick (normally 2 hours daily incremental backup for all onsite jobs), so I don't have the need for parallel processing. I prefer not to saturate the production storage with multiple jobs even if it is capable for more.
* I like to get the simple daily bottom line report that I get from my powershell script and looks this:
FYI My setup is:
A new 3PAR 7200 with FC connections (just migrated last week from HP Lefthand P4300).
3 VMware hosts (HP DL360 G7/G8), esxi 5.1u1
1 physical backup server (Dell R610)
An IBM DS3300 ISCSI Storage for backup repository
Currently we have about 30 VMs to backup,
with a total of about 2000gb production data to backup daily.
As mentioned earlier it takes about 2 hours for daily jobs, using reverse incremental.
Some VMs have also offsite backup (regular Veeam job to remote repository).
Thank you very much for the feedback, I can certainly understanding wanting the jobs split. Based on your description it seems like you could simply set the number of parallel processes on your repository to 1, and get sequential processing without needing to chain jobs at all, just allow the scheduler to queue up the jobs and run them in sequence. You'd still need some Powershell to get your single nightly report though so certainly your method works fine and is quite simple.
