Direkt zum Inhalt

Replication unreliable

Thread needs solution

Hi,

configuration: ACR125 Std Update 1: primary location and first replication location on LAN, second replication is on remote location accessed over link (30Mbit/s).

1.

A backup plan for server (300GB of data) was created and run successfully up to Copying to second replication location over slow link. Copying started well. After few hours it stops with error:

Command has failed. Command=Replication;
Network disconnected.

Yes link is not 100% reliable however I expected that ACR will automatically restart copy (rsync was used before and it works this way).

2.

Next I run backup plan again. It stops just before Copying stage should started with:

Backup failed
The file is corrupted.

Hm, this means ACR was not able continue copying from previous backup.

3.

I delete corrupted backup on remote location and rerun the plan. It stops again with error:Command has failed. Command=Replication;
Network disconnected.

and running again show error:

Backup failed
The file is corrupted.

4.

I delete corrupted file again. Run backup plan and after some time ACR reports it is finished successfully. However the file size on remote location was only about 1/10 of the size of the file on primary location. And there was no error during running backup plan.

I run backup plan again and it finished just before it should start copying to remote location with

Backup failed
The file is corrupted.

 

Conclusion:

It looks like ACR backup replication functionality is unreliable and even un-safe on unreliable links. It is not able to go further from point where it stopped before - this is real trouble for usage on slow and unreliable links. Even more it is at least ones reported that backup is ok, however file was corrupted.

Am I missing something or should avoid this functionality? Somewhere was mentioned that ACR use a kind of transaction oriented approach in creating/copying backup to achieve reliability. With this case this is not proved.

 

 

Simon

0 Users found this helpful
frestogaslorastaswastavewroviwroclolacorashibushurutraciwrubrishabenichikucrijorejenufrilomuwrigaslowrikejawrachosleratiswurelaseriprouobrunoviswosuthitribrepakotritopislivadrauibretisetewrapenuwrapi
Beiträge: 22
Kommentare: 3800

Hi Simon,

There is a known issue related to default backup format (version 12) that replication from such archives does not properly handle certain network dropouts. You can do the following to workaround the issue: create a _new_ backup plan, choose the same backup destination, and switch the Backup Options->Backup Format to version 11. After that replication from the 1st location to 2nd one should successfully re-try the network dropouts. To track when the root cause (issue related to Backup format v12) is fixed, please submit a request to our support team. 

Thank you.

Hi,

let me clarify. Is this issue related only to replication from 1st location to 2nd, 3th? Or also to backup to the 1st location?. So can backup format version 12 be used for backups without replication or it is even in this case recommendation to use backup format version 11 to avoid this issue?

 

 

Simon

 

frestogaslorastaswastavewroviwroclolacorashibushurutraciwrubrishabenichikucrijorejenufrilomuwrigaslowrikejawrachosleratiswurelaseriprouobrunoviswosuthitribrepakotritopislivadrauibretisetewrapenuwrapi
Beiträge: 22
Kommentare: 3800

Hi Simon,

The issue is related to reading data _from_ backup format v12, so writing data during backup into this format can sustain the network dropouts, but not replication or recovery from it for example.

Thank you.

Thanks for this.  I am similarly situated.  If I reboot my home router, that halts the currently running replication(s) and rather than picking up where they left off, they just start over.  Some of my backups are large so just re-replicating from zero is problematic.  I'll try this solution.  Thank you!

frestogaslorastaswastavewroviwroclolacorashibushurutraciwrubrishabenichikucrijorejenufrilomuwrigaslowrikejawrachosleratiswurelaseriprouobrunoviswosuthitribrepakotritopislivadrauibretisetewrapenuwrapi
Beiträge: 0
Kommentare: 2016

Hello Max,

thanks you for posting on Acronis forums!

You are welcome to share the results of applying this solution. Please also do not hesitate to ask questions if you have any.

Maria Belinskaya wrote:

You are welcome to share the results of applying this solution. Please also do not hesitate to ask questions if you have any.

Thanks Maria.  My bad...  Even though I entered into the forums using a True Image 2020 link, I somehow ended up in a Backup 12.5 forum.  I have the exact same symptoms in True Image 2020, but I don't have the means to apply this workaround as I do not have interface control over file formats in True Image.  I've had to revert back to separate local and cloud backup jobs, and not rely on replication.  Slightly less convenient, but so much more control that it really works better for my situation.

frestogaslorastaswastavewroviwroclolacorashibushurutraciwrubrishabenichikucrijorejenufrilomuwrigaslowrikejawrachosleratiswurelaseriprouobrunoviswosuthitribrepakotritopislivadrauibretisetewrapenuwrapi
Beiträge: 250
Kommentare: 7092

Max Roesler wrote:
I have the exact same symptoms in True Image 2020, but I don't have the means to apply this workaround as I do not have interface control over file formats in True Image.  I've had to revert back to separate local and cloud backup jobs, and not rely on replication.  Slightly less convenient, but so much more control that it really works better for my situation.

Dear Max,
I've commented on this limitation here https://forum.acronis.com/comment/516663#comment-516663