Knowledge Base & Discussion Forum

6.11 replication errors.

Discuss technical questions on AhsayRPS

Moderator: Support Team

Re: Helllooo Support? Bueler? anyone?

Postby millhaus » Wed Nov 21, 2012 9:01 pm

obwielnls wrote:148665 06:35:13 PM [Thread][RPSSender][admin] Error=[SendHandler.close] Unable to write close command. Error='Connection has been shutdown: javax.net.ssl.SSLException: java.net.SocketException: Connection reset'
148666 06:35:14 PM [Debug:RPS] Error in unsync mode. Connection reset
148667 06:35:14 PM [Thread][RPSSender][admin] Error=[RPSSender.replicateFileUnsync][Exception] Connection reset
148668 06:35:14 PM [End] Replicating files in UNSYNC mode
148669 06:35:14 PM Sent 1.85G byte in 14,927 sec (130k byte/sec)
148670 06:35:14 PM [RPSSender.replicateFileUnsync][Exception] Connection reset
148671 06:35:14 PM Connection reset
148672 06:36:14 PM [Start] Replicating files in UNSYNC mode


By chance, I had this issue yesterday, too!
RPS starts replicating, logs, from "Refreshing file" (replicates %programfiles%\Ahsay OBS\)
. Now when it should start replicating the UserHomes\UserData, it does not start.
Last entry was
Code: Select all
Replicating files in UNSYNC!


Luckily, I have found my mistake - at least I think I found it- After upgrading RPS from 6940 to 61100 I have overwritten the old backup server.conf etc.
Check your %programfiles\Ahsay OBS\conf\ folder. I.e. the server.xml
After I have used the new server.xml by Ahsay 61100 RPS did continue!
millhaus
 
Posts: 279
Joined: Tue Oct 04, 2011 9:57 pm

Re: Helllooo Support? Bueler? anyone?

Postby obwielnls » Wed Nov 21, 2012 9:36 pm

By chance, I had this issue yesterday, too!
RPS starts replicating, logs, from "Refreshing file" (replicates %programfiles%\Ahsay OBS\)
. Now when it should start replicating the UserHomes\UserData, it does not start.
Last entry was
Code: Select all
Replicating files in UNSYNC!


Luckily, I have found my mistake - at least I think I found it- After upgrading RPS from 6940 to 61100 I have overwritten the old backup server.conf etc.
Check your %programfiles\Ahsay OBS\conf\ folder. I.e. the server.xml
After I have used the new server.xml by Ahsay 61100 RPS did continue![/quote]

Mine actually starts replicating the user data.. Sometimes it makes it 20% through sometimes 90%.. seems to be no pattern to when it quits other than the 15 minute thing.. when it stops, it hangs for exactly 15 minutes then gives the error above.
obwielnls
 
Posts: 166
Joined: Sat May 17, 2008 8:41 pm

6.11

Postby safedata1 » Tue Nov 27, 2012 12:44 am

I too have errors with replication since 6.11.

Was working fine on 6.9.4.0 then stupidly upgraded OBS to 6.11 (I always ask why did I do that after the event)

Replication has NEVER been a strong point of Ahsay. It's fine for any OBS around > 2TB anything larger it's 50/50 and if you ever reach logging mode you know it won't stay there for long and that's not because of it's monthly resync.

My errors are
[Debug:RPS] Error in unsync mode. Read timed out
4918 01:05:53 AM [Thread][RPSSender][reseller] Error=[RPSSender.replicateFileUnsync][Exception] Read timed out
This normally shows after 30 minutes of no log activity. I will try and put RPS back to 6.9.4.0 but I won't hold my breath.

Not impressed (again) with a new release
safedata1
 
Posts: 2
Joined: Thu Nov 03, 2011 11:11 pm

Postby PG » Tue Nov 27, 2012 4:14 am

We had RPS errors since the dawn of time, do yourselves a favour and ditch it ;)
PG
 
Posts: 1023
Joined: Thu Feb 18, 2010 5:57 am

Re: 6.11

Postby millhaus » Tue Nov 27, 2012 3:33 pm

safedata1 wrote:...
[Debug:RPS] Error in unsync mode. Read timed out
4918 01:05:53 AM [Thread][RPSSender][reseller] Error=[RPSSender.replicateFileUnsync][Exception] Read timed out


Exactly the same error! Read timed out.
RPS has reached only 50% of OBS UserData, and it starts again :evil:

Code: Select all
1353974007831,[info],Deleting remote 'L:\UserBackup\<customer>\files\1291191907520\2011-10-18-08-41-57'
1353975809094,[error],[Debug:RPS] [b]Error in unsync mode. Read timed out[/b]
1353975809094,[error],[Thread][RPSSender][<receiver>] Error=[RPSSender.replicateFileUnsync][Exception] [b]Read timed out[/b]
1353975809094,[info],[End] Replicating files in UNSYNC mode
1353975809094,[info],"Sent 56,15G byte in 252.929 sec (232k byte/sec)"
1353975809094,[error],[RPSSender.replicateFileUnsync][Exception] Read timed out
1353975809094,[error],Read timed out
1353975869166,[info],[[b]Start] Replicating files in UNSYNC mode[/b]
1353996009200,[info],RPS Version: 6.11.0.0
1353996058969,[info],Listing remote 'C:\'


I'll dump this "solution" as soon as I have a better and intelligent robo solution. Unfortunately, I am still unaware how to replicate ~ 7TB of data via robocopy.

You cannot run a single robo job per user which would mean ~ 130 jobs simulatenously which would break down our poor OBS disk IO!!! You can neither run one job to replicate ALL LUNs sequentially, robo would take years to list all DAMN SMALL files created by Ahsay!!!

I guess I might create per user (depending on its data) a robo job which will run via task scheduler between morning untill evening, making sure only about max. 5x parallel jobs running with /IPG:<s> in order to reduce CPU- and IOPs load. Damn hard to manage this stuff :shock:
millhaus
 
Posts: 279
Joined: Tue Oct 04, 2011 9:57 pm

Postby Jonek » Tue Nov 27, 2012 11:48 pm

I have had a similar issue in the past on 6.9, and at that point the issue was solved by setting the Java heap to half the physical memory.

Have you changed the memory settings of the OBS?

At the moment, I am having an open ticket regarding hanging of the updated 6.11 server ... memory settings are as they where on 6.9, but it seems as if my replication takes the whole server down (or at least more or less unresponsive)
Jonek
 
Posts: 416
Joined: Fri Jan 04, 2008 7:05 am

Postby millhaus » Fri Dec 07, 2012 12:59 am

Jonek wrote:I have had a similar issue in the past on 6.9, and at that point the issue was solved by setting the Java heap to half the physical memory.

Have you changed the memory settings of the OBS?

At the moment, I am having an open ticket regarding hanging of the updated 6.11 server ... memory settings are as they where on 6.9, but it seems as if my replication takes the whole server down (or at least more or less unresponsive)


As you advised, I have increased heap on OBS; RPS tomcat has already been started with half of physical RAM.

xms96 > xms128
xmx7228 > xms9216

replication is now working since the last two days without interruption error.

One more question:
Is retention only working if RPS won't be interrupted?

Example: retention=35d; RPS did delete some user, even though OBS userhome deletion happened only 14 days ago?

I fear RPS does not keep anything in retention when he does restart replication due to interruption errors?
millhaus
 
Posts: 279
Joined: Tue Oct 04, 2011 9:57 pm

Previous

Return to AhsayRPS

Who is online

Users browsing this forum: No registered users and 2 guests

Looking for Rbackup Alternative | Vembu Alternative | Novastor Alternative | Asigra Alternative | BackupAgent Alternative? Try our product.


A wholly owned subsidiary of Ahsay Backup Software Development Company Limited  [HKEx Stock Code: 8290]