size backup proxy

[crazych]

Hello please say me does the size backup proxy on perfomance(IOPS) my storage

[Vitaliy S.]

Hi Roman,

Can you please clarify your request, as I'm not sure I follow it? Do you want to know what CPU/Memory configuration do you need for the backup proxy? As for the IOPs then it depends on the backup mode and the number of jobs you're currently running.

Thanks!

[crazych]

No, now as the proxy use a separate VM to use a proxy folder on the disk size of 4 gigabytes. Whether this is enough for normal operation of veeam backup? The configuration of the VM 4 GB of RAM and 4 cores CPU. Recently noticed that the backup speed dropped to 25 megabytes per second this creates a heavy load on storage

[Vitaliy S.]

It is expected to have load on the repository storage during backup job run. Veeam backup proxy storage does not matter here. What does bottleneck stats show you in the Veeam console?

[crazych]

:: Load: Source 97% > Proxy 17% > Network 10% > Target 5%

the backup speed 5-6 megabytes per second. Storage based on FreeNas 9.3+ iSCSI as the target is used 2 disk in mirror and normal linear reading even inside of a VM to 50 megabytes per second

[Vitaliy S.]

Seems like the source storage that hosts VMs cannot provide more IOPs for the backup job. What backup mode (SAN, hotadd, network) do you use?

[crazych]

I also think the problem is in the repository just can't understand where the problem is. I use as a backup destination CIFS network share

[foggy]

The problem is not in the backup repository that holds backups, but in the source storage (production) where the backed up VMs reside. Please refer to the job session log and look for the transport mode tag after the proxy server name (should be one of the following: [san], [hotadd], [nbd]).

[crazych]

for some reason disk 1 is copied slowly, the second quickly, they are both in 1 place. May be corrupted VMDK file?

[Vitaliy S.]

Based on the screenshot, I can see that you had a failover to network processing mode for the 1st disk. Can you please tell me what disk type is that? Just to confirm - do you have both disks on the same datastore or not?

[crazych]

Both drives are on 1 storage on 1 LUN . Partition LUNs formatted with VMFS5.60 file system ntfs VMDK disks

[veremin]

Does any of those apply to the first disk? Thanks.

[crazych]

yes
Hot Add may fail if the VM you are trying to back up and the proxy server are in different clusters.
backup server is in a different cluster
this is a problem for 1 disk VM?

[crazych]

The transfer server in the same cluster that the VM is reserved and that did not solve the problem. Could it be the problem is that the VM hardware version 10?

[crazych]

used when hot add the disk should be visible in the list of drives with an assigned drive letter on the backup server? Because at the time of the backup VM, I see the drive in windows Explorer and can navigate change.
here you can write in Russian?

[foggy]

Drive letter assignment might occur during hotadd, please check whether automount is disabled on the proxy server.

While these forums are for English speakers only, you can contact technical support to get assistance in your native language.

[crazych]

After you remove disconnected drives from device Manager, the drives have become connect as "Using backup proxy VMware Backup Proxy for disk Hard disk 2 [hotadd]"
But the backup speed for the same low

[Gostev]

Your issue looks to be due to poor random I/O performance of your primary storage. If you disagree with this conclusion, feel free to open a support case to let our support staff demonstrate the issue using industry standard disk I/O performance testing tools. Thanks!