"X GB read" in the log file

[brianlees]

Am I correct in my understanding that the "x.x GB read at x speed" in the log file refers to the amount of actual changed blocks, correct?

We are seeing some seriously high numbers for a backup that is occurring on one server every hour that give me pause for concern....the app, not Veeam. I do have exclude the swap file checked in the job, so we are fine there.

[brianlees]

Hmm... I noticed that "Swap file blocks skipped" is shown on the full backup log, but not on the incremental logs. ?????

[Gostev]

1. Correct.

2. Either the swap file is not changing, or small changes happen in blocks that also include other files, so we cannot skip them (system managed swap files are usually pretty fragmented).

[brianlees]

Hmm... So there is a possibility that it is grabbing swap files? That makes sense. I guess we need to go back and correct that with fixed sizes.

[veremin]

As mentioned, it might be possible that VB&R grabs swap file blocks that also contain other files.

However, the high amount of changes is typically related to defragmentation or antivirus activities that run inside VM on regular basis. Also, special applications such as SQL, for instace, are notoriously known of producing high change rates.

Thanks.

[brianlees]

Some of the servers are SQL and Oracle. The one Oracle one has like 1.5 to 2gb change per hour (I did an hourly backup test yesterday), which seems crazy as we don't use it nearly that much. That is why I was thinking some swap file issues as well.

[brianlees]

I'm still going to go through and change all our servers away from system managed page files, though. Just makes sense from a Veeam perspective.

[foggy]

If you're concerned with the amount of changes in the guest OS, then this topic could probably give you some hints.

[brianlees]

Thanks. I just checked a few servers (all are based on the same template) and they are all disabled for the LastAccessUpdate. But, that was a very good suggestion to check!