Is Replication not Ready for "prime time"???

[pcmmitham]

I get the disticnt feeling that the "replication" portion of Veeam B&R hasnt been fully tested. There are soo many little issues with this function.

1)if you change any of the options after you create a job, it will fail and you have to rebuild the job
2)default settings for replication are completly wrong resulting in needing step 1
3) contsant errors even when replicating to another device on the LAN. If it cant do it on the LAN how will it ever be able to do it on the WAN???

i'm a bit frustrated with this. I just rebuilt all my replication jobs (again) and now I'm getting the following error:

Error: Skipping initial replica synchronization: seeding is enabled

this is on the second run of a replication job that previously succeeded all it's VM's doing their inital seed and replication! How can an already seeded and replicated VM be skipped because seeding is enabled?? makes no sense!

[Jfmoots]

Pierre,

I don't think your situation is the norm, but more the exception. My experience with replication has been quite positive. A forum like this can give a false impression that everyone is having problems with a function, but it's important to consider that people don't often come and post "works great! no problems!" Replication can get very complex and there's lots of "bumps" to overcome to get a VM from one host to another and then keep it updated. In a Eutopia, we work perfect, but as you add hurdles we're certainly going to need adjustment. Every environment is unique.

The improvements we've made over v5 have been well received and increased performance. I have many new installations with replication working flawlessly, including our seeding, re-ip, and network mapping features. I also have lots of installations the were upgraded from v5 where the performance gains and added features have been clear.

I understand your frustration and I'd recommend you open a support case regarding your error. That seems odd and I suspect there's a problem with that replica.

[foggy]

How can an already seeded and replicated VM be skipped because seeding is enabled?? makes no sense!

Is there a chance that this VM was recently re-added to the inventory?

[pcmmitham]

no, this is the exact same job (un altered in any way) that I ran a few hours ago!

as to the other setup issues, they've all been behaviours confirmed by veeam as normal. I really like the Veeam product,it's far superior to vDR or similar technologies. but not being able to adjust settings after you create the job is just silly!! after all how am I going to make adjustments to things once I move the DR server offsite and really start replicating! I understand that the forums are generally where the "whinning" occurs.

[lobo519]

I have been using replication since version 5. It has work great for me. And as stated - v6 has brought many welcomed improvements. So much infact - we were able to eliminate we WAN accelerator.

In addition - I have changed my replica jobs several times - things such as the host to replicate to, the datastore to replicate to, the proxy it uses. I have used the mapping function in most of these cases. Never once rebuilt the job.

[Gostev]

I get the disticnt feeling that the "replication" portion of Veeam B&R hasnt been fully tested.

I would guess this has something to deal with issues in your own environment, because feedback from everyone else proves the opposite. Here is the thread from almost half a year ago - shortly after v6 release, even before we released any patches... I have picked such an old thread intentionally to address your quality of testing concern > v6 WAN Replication - Wow!

v6 replication is rock solid even with the RTM code, and is currently working in production of:
- One of the largest European bank (hundreds of sockets)
- One of the largest European retail chain (hundreds of sockets)
- One of the largest US cloud providers (in the beginning of major project)

I can definitely say that replication is one thing in v6 that I am totally confident about.

[foggy]

no, this is the exact same job (un altered in any way) that I ran a few hours ago!

I mean was this VM maybe removed and then added again to vSphere inventory?

[pcmmitham]

Again,

This job is unaltered in any way, the vm's are unaltered in any way.. Literally I ran the replication job, then a few hours later (with not change to either veeam or the VMware environment) ran the job again as would be normal and got this error.

Gets even more interesting today. One of my other jobs ( after successfully seeding, and 2 subsequent successfull replications) suddenly decides that one of the vm's in the replication job must be skipped as "seeding is enabled". WTF??? The vm has already been seeded and replicated over the LAN why does it need seeding again.

Oh well maybe replication will be fixed and ready for production with the imminent release of version 6.1!!!

I'd hate to think of the issues I might be running into if I had a complex environment. I only have 3 hosts and 34 vm's

[foggy]

Pierre, we definitely need to review the logs of these jobs. It would be great if you could open a case.

[Berniebgf]

We have just finished failover and failback testing for a customer, 40 + VM's replicating.
Full failover and failback test to DR site, Veeam 6.0 with updates applied supporting VMware 4.1 and 5.0... Worked well, wish I could say the same for our physical servers@!

[pcmmitham]

Foggy, I already have a ticket open (a few actually). The answer for everything seems to be the same... Recreate the job, frankly that answer is unacceptable. I can't keep bringin my dr server back to head office every few days to reseed jobs ( replica mapping is hit and miss)
Ticket 5179323