Replication Seed Failing

[matt19849]

Hi,

I am trying to set up a replication job using a seed. One of the VM's has gone through with no issues but the other is failing.

The error I am getting is:

Server01 Success 13:28:16 14:51:16 3.7 TB 2.6 GB 1.7 GB 1:22:59
Server02 Error 06:54:46 06:56:34 0.0 KB 0.0 KB 0.0 KB 0:01:48 Processing configuration Error: Cannot complete the operation because the file or folder [Veeam Backups] Server02_replica/Server02_1.vmdk already exists
Error: Cannot complete the operation because the file or folder [Veeam Backups] Server02_replica/Server02_1.vmdk already exists
Server02 [seed] Success 13:28:18 06:54:42 (+2) 6.9 TB 5.9 TB 5.7 TB 41:26:23

The file didn't exist until the seed created it. I have tried to remove all files and folders related to this replication and start from scratch again using a seed and I get the same error. The replication creates the folder from the seeding part of the job. This goes through fine, it then fails on processing that server to check for changes etc with the above error

I have a case open with Veeam but so far I haven't had an answer that works and this is becoming a bit of an issue for us

Any help would be appreciated

Cheers

[Gostev]

Hi, please include support case ID as requested when you click New Topic. Thanks!

[matt19849]

Hi, please include support case ID as requested when you click New Topic. Thanks!

Apologies: 00624300

[Shestakov]

Hi Matt,

...Server02 Error 06:54:46 06:56:34 0.0 KB 0.0 KB 0.0 KB 0:01:48 Processing configuration Error: Cannot complete the operation because the file or folder [Veeam Backups] Server02_replica/Server02_1.vmdk already exists
Error: Cannot complete the operation because the file or folder [Veeam Backups] Server02_replica/Server02_1.vmdk already exists
Server02 [seed] Success 13:28:18 06:54:42 (+2) 6.9 TB 5.9 TB 5.7 TB 41:26:23

Did you try to seed the same server twice?

[matt19849]

This was the first time I have seeded that server, the initial seed went through fine but then failed. I hit retry and it failed again

[Shestakov]

Matt, could you elaborate on this:

This was the first time I have seeded that server, the initial seed went through fine but then failed.

What do you mean by "the initial seed went through fine" ?

I`m trying to understand where in the replica seeding process the error is.
If Veeam B&R created a replication job and synchronized the restored VM with the latest state of the original VM, it should not restart the current job or start a new one.

Are you able to you see a VM replica with two restore points in the Backup & Replication view of Veeam Backup & Replication, under Replicas node?

Another option is to try to Seed a Replica Manually.

Thank you.

[matt19849]

Sorry, I may not be explaining this very well.
When I created the job i selected the option to seed from a backup file that was stored on another datastore. The job started to go through fine but seemed to fail on the second part of the first replication.

http://jpaul.me/wp-content/uploads/2011 ... ighted.jpg
That picture isn't of our environment or our replication but it shows what I was seeing, in the picture the one with [seed] went through fine but the other below failed with the mentioned error

[Shestakov]

Thanks for the clarification, Matt.

The job started to go through fine but seemed to fail on the second part of the first replication

During the first run of a replication job, Veeam B&R accesses the repository where the replica seed is located, and restores the VM from the backup on the replication target host in the DR site. Files of the restored VM are placed to the location you specify as the replica destination storage.

So, you can check if this stage was done by checking if you can find VMs on the replication target host.

Have you tried to try to Seed a Replica Manually?
It`s great to know where exactly stumbling block is. Thanks.

[matt19849]

Hi,

The VM is in the location where I have asked it to be.
I have tried to map the replica to the VM that was copied over from the seed but it still errors with the same error

[Shestakov]

Ok. Let`s go step-by-step. For a replica seeding we need to:

1. Create a backup of a VM that you plan to replicate.
2. Copy created backup to the backup repository in the DR site.
3. Create a replication job and point it to the backup repository in the DR site.
4. Veeam B&R synchronizes the restored VMs with the latest state of the original VMs.
5. For all subsequent runs of the replication job, Veeam B&R transfers only increments.

You`ve said you have the VMs on the replication target host, means step 3 is done.

To check step 4 status, let`s go back to my question: "Are you able to you see a VM replica with two restore points in the Backup & Replication view of Veeam Backup & Replication, under Replicas node?"
One point contains the state of the VM from the backup file, another contains the latest state of the original VM you want to replicate.

[matt19849]

Hi,
Just the one restore point is showing

[Shestakov]

Means the VM has a restore point only from the backup file.
The good thing is now we know where the process stopped.

Please, follow the instruction from the part 2 to finish replica seeding.
Let us know if it works.

Thank you.

[matt19849]

Hi,

Sorry for the delay in responding. I have tried following the instructions from step 2 but I still get the same error

Cheers

[Shestakov]

Matt,
That`s unexpected behavior, logs need to be reviewed.

Seems like DR site still keeps some orphaned files which VBR tries to use. I would try to carefully check DR for the old snapshot files.
If it doesn`t help it makes sense to change "Server 2" name and repeat replica seeding from the scratch.

But since you have already provided logs to Veeam Support, let`s see what they will say.
Thank you.

[matt19849]

No problem, I have tried to start the replica from scratch but get the same issue as before
I'll wait to hear back from Veeam

[Shestakov]

Matt,
Just curious, did you change the name of the server before trying to make replica seeding from the scratch?

[matt19849]

I didn't change the name on the second time around - that is worth a try though

[Shestakov]

Matt, yes, it`s interesting if it would work.

Let us know the result of the logs reviewing. Thank you.

[matt19849]

Just a quick update on this, I still have no valid response from Veeam and seem to be going around in circles with them at the moment.
I have tried to re-create the job using a different replica name i.e server_replica01 but the same error occured

[Shestakov]

Thank you for updating the post, Matt!

As I see that support team asked you to do same things that were suggested here(because it usually helps).
I hope upper level of support will help you to investigate the issue.

Please post again once you find out the problem. Thanks!

[nunciate]

I am also having this issue after having some failed replications due to latency on my SAN. I have 3 VMs that have been replicating for many months with no errrors. The jobs failed last night and now when I re-run the jobs I get the follwoing message.

9/12/2014 6:31:03 AM :: Processing configuration Error: Cannot complete the operation because the file or folder [NA03_7200_LUN08] TXPLVS228_replica/TXPLVS228_1.vmdk already exists.
The seed option is checked and the seed tab has the replica mapped. This setting has always been there so I didn't change anything. I keep 2 restore points and I only see those 2 snapshots on the replica VM. The prod VM looks fine as well.

It seems like every time I have a failure I end up having to reseed the entire VM out there to get it synced up. I think Veeam needs better error handling when it comes to replications. I have many VMs that are very large and reseeding the entire thing is a long task. I had a 10Tb file server that we replicated with Veeam and it failed a while back and took me over 2 months to get it fully back online in DR.

Case # 00635506 has been submitted.

[Vitaliy S.]

Hi Alan,

Have you tried just re-mapping your replication jobs in order not to do the full seed again? Does it work?

Thanks!

[nunciate]

I deleted the entire job and recreated it from scratch mapping the replica to the job and that didn't work. Same error.

[veremin]

Then, please, keep working with the support team. The root cause will be found, once required logs are analyzed. Thanks.

[Amol]

Hello,

I have been receiving an error message during an replication job, error message is "Cannot complete the operation because the file or folder already exists". I can't understand what it is all about, anywhere i should look into before i log a case for this.

Thanks,
Amol

[veremin]

I assume there has been a failure happened during replication cycle (outage, storage problem or something) and now your replication is out of sync. Please, open a ticket with our support engineers and let them address it directly. Thanks.

[Amol]

Hello,

I have been receiving an error message for an replication job, error message is as follows "Cannot complete the operation because the file or folder already exists". At first, it was only one replication job which was affected but later second job also started facing the same. Have also logged a case with support, case id is 00932778. What they suggested is to map and seed the replica, which i did and it seems to have worked for first job but the second job still continues to face the issue. What is it now i should look into? Thanks.

Regards,
Amol.

[Shestakov]

Hello Amol,

What they suggested is to map and seed the replica, which i did and it seems to have worked for first job but the second job still continues to face the issue. What is it now i should look into?

Did you do anything differently for the second job?

You can also try to make those easy steps manually as I described above in the thread:

For a replica seeding we need to:
1. Create a backup of a VM that you plan to replicate.
2. Copy created backup to the backup repository in the DR site.
3. Create a replication job and point it to the backup repository in the DR site.
4. Veeam B&R synchronizes the restored VMs with the latest state of the original VMs.
5. For all subsequent runs of the replication job, Veeam B&R transfers only increments.

Here is also a description in the related KB. Thanks!

[Amol]

Hello Shestakov,

Have done this as support team also suggested the same, but still receive the same error "Cannot complete the job because the file or folder already exists". I am really concerned as we have a scheduled DR cut over in next two days and with no replication jobs running its gonna be a bad day.

[foggy]

Amol, you an raise the severity of the issue and I believe support will be able to address it promptly. I would also try to re-create the job and perform seeding from scratch.