9.2 and Server Share Interruption while replicating?

[jwojda]

Has anybody noticed any problems with server share interruptions while replicating since going to 9.2?

It goes through the replication and will just randomly stop. a dialog box will pop up and said server connection interupted and I can either ignore or disconnect. If I choose ignore C.Admin idles. (server app will show no activity) and if I choose to disconnect C.Admin idles. I can only force quit C.Admin and manually drop the connections from my desktop and force disconnect from Server.app.

Originally I thought it was a speed issue across the wan, but I just tried a machine over gigabit and it bombed out as well.

This particular case is C.Admin launched from my machine which is 10.9, then it mounts the main DP which is 10.8.5, and the target server is a 10.9... maybe it's 10.9.
*sigh*

[chris_kemp]

Hmm...this sounds suspiciously like an issue I'm having right now - however, it's with 8.71 under 10.8.5. I am seeing the same thing, though - the replication proceeds normally at first, and then drops the connection to the main server share. It does not close cleanly, either - I basically have to reboot to force removal of the share connections.

I've been troubleshooting the networking, but so far have not come up with any concrete issues...

[Chris_Hafner]

I'm seeing this as well... or at least I think I am. I'm running the JSS (9.22) on a 10.8.5 box, replicating to our secondary DP, which is also the SUS and hence, is running 10.9.1 with server 3.whateverscurrent. I'm using a 10.9.1 iMac with casperadmin to replicate. When the connection drops the unit wasn't to automatically replicate ALL packages from on to another. I haven't dug into it yet, but wanted to mention that I'm seeing some weirdness like that as well.

[SamF]

I've seen a similar issue which has been exclusive to 10.9 shares and found the underlying cause to be related to the network connection getting dropped entirely. This discussion outlines the behavior we saw with Mavericks boxes hosting the CasperShare:

https://discussions.apple.com/thread/5483424

[Chris_Hafner]

Thanks for that. A weird bit to add. In pre-testing yesterday I was magically able to replicate ~350 gigs from our Master DP (10.8.5 server 2.x) to the Secondary (10.9.1 server 3.x) via C.Admin 9.22 on my 10.9.1 iMac. It's weird, but I will begin playing with the script that was recommended in the thread you posted Sam. Thanks for that one, it's a good one!

Looks like it may or may not exactly be the cause of our issues. Unfortunately, our department is also in the midst of trying to introduce Juniper switches to our outdated 3com/HP infrastructure in places and there are... hiccups. So, unfortunately, it's very difficult for me to trace at this exact moment in time. That... and some sprinklers bursting in several classrooms taking out a swath of POE network devices. *loves winter*. In any event, if anyone else notices this please post your experiences!