cfrost113 Posted November 8, 2010 Report Share Posted November 8, 2010 This seems to happen on all macintosh workstations that we have configured for. They are all running the latest version of the Retrospect client (6.0.329) on MacBook Pro's running OSX 10.6.4, and the server runs version 7.7.341 (64bit) on Windows Server 2008 R2. It seems that when doing either a proactive or a scheduled script for each Macintosh (they have their own backup set for each laptop), they start ok and get some data then fail. Here's an example output from the log of one laptop: + Normal backup using %username%'s Macintosh Backups at 11/8/2010 2:30 PM (Execution unit 1) To Backup Set Mac Backup Set A... - 11/8/2010 2:30:19 PM: Copying Macintosh HD on %username%’s MacBook Pro Trouble reading files, error -519 ( network communication failed) 11/8/2010 2:49:43 PM: Execution incomplete Remaining: 30232 files, 18.2 GB Completed: 264 files, 1.1 GB, with 0% compression Performance: 110.9 MB/minute Duration: 00:19:23 (00:09:32 idle/loading/preparing) To mitigate this so i can get an entire backup set to start from, i set the retry timeout to 1 minute and all of the other laptops kept doing this error, but i eventually accumulated all of their data. I was able to get my initial first full backup from them. One of the laptops i have still not been able to get a full backup only because this error constantly comes up so often that it crashes or freezes the "retroclient" process on the MacBook Pro it's running on. I have even seen it kernel panic because of this. I verified the entire time that network connectivity was ensured. There is only 2 switches in between the server's NIC and the client's NIC's. Basically, the layer 2 would be from Server NIC, to switchport on server switch 1, through trunk to distribution switch, to switch port of the client, through a Cisco Voip phone, to the client workstation. I even tried without the Voip phone on the same switch and still got the network timeout values. I took all spanning tree configurations out and any QoS configurations out also. This just makes no sense now, so any help is greatly appreciated! Quote Link to comment Share on other sites More sharing options...
blm14 Posted November 9, 2010 Report Share Posted November 9, 2010 The best thing to do here is to kill (from command line) the retrospect client processes (all of them) on the client. Restart the retrospect client (again, from command line): /Applications/Retrospect Client.app/Contents/Resources/retroclient -log 9 I think is the switch to enable the highest level of logging. Then attempt to do a backup job and if it crashes then you should open a ticket with retrospect and send them the logs from the client which are in /var/log/retro*.log Quote Link to comment Share on other sites More sharing options...
cfrost113 Posted November 12, 2010 Author Report Share Posted November 12, 2010 Thank you for that, I will try this and I'll post my findings on here for all to see Quote Link to comment Share on other sites More sharing options...
cfrost113 Posted November 15, 2010 Author Report Share Posted November 15, 2010 I have done some optimization of my network adapter's settings on the server just to see if this had any effect. I am running an Intel Pro 1000/EB network adapter on a S5000VSA motherboard. The main issue that i was having was a TCP/IP problem. We were getting a lot of TCP Zero Window packets coming from the clients because they were receiving too much data from the server. The server was basically not waiting for the clients to be done processing packets and send another window to receive more packets. I actually turned off all flow control of any Tx and Rx packets on the server and this mitigated the issue somewhat. Also, I turned off TCP Checksum Offload and IPv4 checksum offload, because for some reason Wireshark will get these as error packets before the NIC has a chance to process them, so i turned it off to clean up the packet capture a bit. This could have affected the performance of the CPU on the server somewhat, but since i'm running a Xeon 5570, i'm not worried :X Has anyone else seen this issue? Please post your thoughts and any experience you might have had with this error. Thanks! I was able to back up a client Macintosh with no issues of timeouts or anything. Got a full backup out of the first try. Quote Link to comment Share on other sites More sharing options...
cfrost113 Posted December 13, 2010 Author Report Share Posted December 13, 2010 Still getting this issue after adjusting various TCP/IP settings, and reverting to default settings. I am also getting this issue from all of my Windows Servers/client workstations that are on the proactive backup scripts. Any solutions suggestions? Does anyone even listen to these forums anymore? Did Roxio forget people still use/buy this product? Quote Link to comment Share on other sites More sharing options...
aolivera Posted January 13, 2011 Report Share Posted January 13, 2011 I too been having this issue but with only 1 laptop client. Weather its a proactive or a manual backup. In starts off just fine and about 10-15 minutes into it, it fails with "error -519(network communication failed). I moved to a different port on the switch and get the same results. Any ideas? Quote Link to comment Share on other sites More sharing options...
blm14 Posted February 2, 2011 Report Share Posted February 2, 2011 Could be any number of things. 519 is one of the hardest things to diagnose. Is the laptop connected via wireless? if so, a momentary loss of network connectivity would cause that. Quote Link to comment Share on other sites More sharing options...
Daniels Posted February 4, 2011 Report Share Posted February 4, 2011 I would also make sure that the computer and/or hard drives do not go to sleep as this can also lead to -519 error. Quote Link to comment Share on other sites More sharing options...
DWitteried Posted September 5, 2014 Report Share Posted September 5, 2014 I have been messing with Error 519 for awhile now. Tried WiFi and Ethernet (multiple ports on router and switch). I just tried implementing the no-sleep option for the HD (advanced power management, delete time to sleep and it will become Never). I had it set to 999 before. Anyway, got this error which is strange... + Normal backup using Janice-Laptop at 9/5/2014 1:28 PM To Backup Set My Book 3TB - 1... - 9/5/2014 1:28:40 PM: Copying Janice (C:) on Janice_Laptop Can't back up COM+ Class Registration database, error -519 ( network communication failed) Can't back up WMI Repository, error -519 ( network communication failed) Can't back up registry, error -519 ( network communication failed) 9/5/2014 1:41:32 PM: Snapshot stored, 39.5 MB 9/5/2014 1:41:45 PM: 3 execution errors Completed: 9021 files, 14.5 GB Performance: 1696.5 MB/minute Duration: 00:13:04 (00:04:21 idle/loading/preparing) - 9/5/2014 1:41:47 PM: Verifying My Book 3TB - 1 9/5/2014 1:44:28 PM: Execution completed successfully Completed: 11594 files, 14.6 GB Performance: 5686.1 MB/minute Duration: 00:02:41 (00:00:04 idle/loading/preparing) Previously I was getting lots of errors for SkyDrive/OneDrive which I eliminated by disabling the OneDrive feature using the Windows Group Policy Editor (used option two in this link). Quote Link to comment Share on other sites More sharing options...
DWitteried Posted November 6, 2014 Report Share Posted November 6, 2014 I am beginning to believe that the issues I have been having may be due to my router. When I run backups using a cross-over cable between the two computers, or an old router I had lying around; the backup completes with no errors. With the problematic router between the two computers the backups occasionally complete ok, but usually have the errors. Time for a new router when finances permit, it seems to work fine for everything else. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.