*Bug Report* Mac Clients don't execute scripts after OS X Mavericks upgrade (10.9.4) in Retro 11.x

[henry-in-florida]

Hi All, 

 

I've had pretty good success with Retro until recently doing backups but have noticed this error creeping in.

 

After a recent (yesterday) mass upgrade to a bunch of (iMac, MBP, Mac Pro) clients, I noticed a failure of the clients to run their previously working respective scripts. Here is the typical error (same error code on each one!): 

+ Normal backup using Daily iMac at 7/6/2014 (Activity Thread 1)
To Backup Set iMac Backup...
Can't access volume Macintosh HD on Jane's iMac, error -530 ( backup client not found)
7/6/2014 22:31:36 : Execution incomplete

 

The clients are all Retro version 11.0.0.191

 

My workaround was to refresh every one of the clients and begin the backups again. Seems to me that since the clients CPU's run continuously, they were showing up on the source pages as active and were discoverable that this should not happen. So I consider this a bug. Please let me know if anyone has the issue and if there are any fixes for it.

 

Have a look at the rest of my system details in the signature line below. If I can help provide any other information to squash this bug, please advise. 

 

Regards,

 

Henry 

[Maser]

I have not seen a similar problem after upgrading my clients/engine machine to 10.9.4 -- but I add all my clients via hostname (rather than by discovery).   How did you add your clients to the engine?

[henry-in-florida]

I have not seen a similar problem after upgrading my clients/engine machine to 10.9.4 -- but I add all my clients via hostname (rather than by discovery).   How did you add your clients to the engine?

 

Thanks for your reply, Maser. I added most clients using Discovery. Prior versions, I had to add by host IP but version 11 seemed OK in that regard up to now (it found the clients). So didn't think of that  as an issue, but will try to re-add using hostname (IP), since many clients are giving me the same indication albeit randomly. Do you think that simply Locating them (as opposed to deleting, adding the new client, e.g., redoing the script) can accomplish this correctly? 

 

Is this a known bug, and fix?  

 

Henry

[Maser]

I don't know if this is a bug or not (as I've only ever manually added static-DHCP-assigned clients by IP address/hostname...)

[henry-in-florida]

I have not seen a similar problem after upgrading my clients/engine machine to 10.9.4 -- but I add all my clients via hostname (rather than by discovery).   How did you add your clients to the engine?

 

I don't know if this is a bug or not (as I've only ever manually added static-DHCP-assigned clients by IP address/hostname...)

 

I believe that it is and haven't heard any other feedback about it as yet from whomever is gathering bug reports. Is there another place to post them?

 

Added clients Direct IP assignments as Maser did, and haven't had further troubles. All clients added by Direct method (instead of Multicast) do not drop the error. Will keep in mind to add future clients manually.

 

I do consider this at best a temporary workaround, the issue should be addressed as a bug by Retro developers (seems a long standing issue?), so clients added by ANY method (Discovery, Direct IP or by network name) do, in fact work, and continue to do so. Apple's discovery works just fine to recognize network clients, why shouldn't Retro's? 

 

In my case, I have a wireless router that passes through its group of clients, to an internet provider's router. I logged into the router and have assigned specific IP addresses there for clients and server. Discovery is an important feature in Retro. It should work and it is unreliable at this time. Hence this BUG REPORT. 

 

 

Henry

Edited July 28, 2014 by henry-in-florida