[Feature Request] mailbox processing & data size display

[jnana.dash]

Hello,

We just checked the below features are missing in the VBO which will be very helpful for the backup administrators. Hence requesting for feature requests

1. Show the total backup volume size : We schedule various backup jobs with different count of mailboxes. Its always good to show the backup size post discovering the mailboxes, so that we can estimate the time remaining, how much completed etc.

2. Show the size of each mailbox : Some times backup runs with dead slow speed and we dont have any clue whats going on. if the size of mailboxes are visible, at least we can know its because of huge size or high number of items in a mailbox.

3. Show "processing" only for the mailboxes currently being backed up under a job : When a backup job is started, it shows discovery of all mailboxes and then it shows all mailboxes as "processing". We really dont know which all mailboxes are under processing and whihc are under queue. It only displays "completed" when a mailbox backup is finished. Till that time all mailboxes are shown as "processing" and the backup window for each mailbox is calculated form the backup job start time and the mailbox backup finish time, where as the mailbox might be in queue and might not have taken the entire window for the backup. Sometimes the backup is dead slow and takes infinite time during some point of time in a specific job. We dont have any clue in checking which mailboxes are being backed up and what is the issue ? if we know which mailboxes are being backed up, we can check the mailbix size, number of items or in case there is some issue on the mailbox etc. and can get some idean why backup is slow and how much time it gonna take approximately.

Regards,
Jnana Ranjan Dash

[Mike Resseler]

Hey Jnana,

First: Thanks for the feature requests. These are noted.
However, I would like to shed some light on why we don't have this today. In order to be faster, we don't calculate those things. Instead, we immediately start backup of objects/ items once we have the change. Waiting until we enumerated everything, and figured out what the changes are would take a very long time and costly cycles to download would be lost.

I hope it makes some sense
Mike

[jnana.dash]

Hello Mike,

Your point is absolutely valid. But to know the figures is also quite necessary from a backup administrator point of view. I would say, Backup should immediately start after sensing the change ( as you mentioned), but in the back ground it can calculate the size and make the data visible after some time. that should be a win-win situation for both performance and progress visibility.

because without the progress bar, the backup admin is handicaped and just have to wait and watch for the backup completion without knowing an ETA.

Regards,
Jnana Ranjan Dash

[Mike Resseler]

That might be a possibility and something we can discuss internally. Thanks for all, really appreciate it that you think along with us

[palmito]

Hello,
Currently working on a design for a customer with a huge amount of data in o365 (>100TB ExchangeOnline (>10000 users), >100TB SharePoint (>10000 sites), >100TB OneDrive)
we are quite worry about the time that will be needed to make at least the first RUN, even with a quite short item level retention. (1 or 2 months)
Any tips or idea to get more confident or only patience.
Do we have to except several months to perform the first backup before leveraging incremental ? Should we reduce the scope of users during the first year ?
Are there use cases with a lot of users to compare ?

Thanks.
Regards,
Eric

[Johan M]

Hi Eric

We have a customer that is around 1PB that we started more than a month ago, the first full was around 1 month for us, but you should make around 50 application accounts and disable exchange throttling (https://www.veeam.com/kb4198).
Also use a lot of proxy servers and not just 1 and split the jobs out on those proxys.
Can you explain why you use item level and not snapshot based retention ?
On all the new customers we get we only use snapshot based.