Case #07788040 — Problems with Copy jobs.

[1dna]

Hello Forum,

While waiting for the support to escalate i figured i wanted to try write here.

we have our main hq

Site A:

Fiber channel proxy -> Backup Jobs (Immutable 7 days -) Veeam hardend Repository (latest iso) on a Alletra 4140 (Latest firmware)

Site B:
Supermicro server (windows 2025) - 1 big raid on raid 60 (approx 1PB)

We had copy jobs running from SiteA to SiteB for a while - and out of the suddenly they have stopped working
Nothing is going on, just stuck on usaually .nvram

Summary:

Backups from Storage snapshots works fine to the hardend repo - for immutability.
Copy jobs dont want to start (even if there is verified connectivity)

The network is purely Layer2 so nothing between to filter.

We are running MTU 9000 through the entire chain.

So maybe - im just writing here to hope someone might had an idea or tried it yourself.

[david.domask]

Hi 1dna,

Thank you for sharing the case number and the details, and sorry to hear about the challenges.

Looks like you and Support have identified that frequently "unstable connection" is appearing in the jobs / logs, and I believe you're aware of this KB.

I read in the case that a few discussions on firewall / AV have already happened, but for some additional clarity on this error, it's not simply thrown when the connection is interrupted, there is a process that happens when the datamovers detect that data transmitted from the source datamover did not receive an acknowledgement from the target datamover (I am heavily summarizing the exact process). Point is that typically this message means some interference somewhere in the connection, though admittedly it can be tough to pin down.

Ping / Test-NetConnection likely won't reproduce or catch the issue as from my experience it's usually specific data getting false-flagged by whatever is causing the interference.

Please continue with Support as they will review the environment more carefully and there are a few methods of further isolating there the traffic likely gets blocked, but mostly want to reiterate based on my experience that "unstable connection" is a fairly reliable error indicating something interfering, and Support will assist in understanding what that something may be.