CC4 borked?

[andyturpie]

Hi guys and gals,

Since mid December our CC4 install seems to slowly degraded to a point now where we cannot build machines keeps failing at (installing additional packages), my connect is now showing the majority of users as unknown and its currently taking users around 5-7 mins to log in.

Christmas eve, the power was shutdown for electrical work (which we were not privvy to the info), it was an SMS that alerted me from people trying to remote in. The 27th I came in to restart the servers and they shut down almost instantaniously due to the sockets that the UPS where plugged into tripped when the power was restored on Christmas eve!

3rd line and RM have been looking at this for a week and we are still no further forward.

Anyone else suffer from this over sensitivity with CC4?

[X-13]

Since mid December our CC4 install seems to slowly degraded to a point now where we cannot build machines keeps failing at (installing additional packages), my connect is now showing the majority of users as unknown and its currently taking users around 5-7 mins to log in.

I HAD THIS EXACT SAME PROBLEM!!

IIRC, it was something to do with the system account [It has a seemingly random name] being deleted or corrupt.

If you go into RMMC, can you open the general registry settings or user properties for a specific account?

[rad]

LOG A CALL WITH RM!!

They are there to help you, log it, and escalate it if you need to.

[X-13]

LOG A CALL WITH RM!!

They are there to help you, log it, and escalate it if you need to.

From the sounds of it, he has.

This looks like Andy is just asking to see if anyone knows what it is.

[andyturpie]

I HAD THIS EXACT SAME PROBLEM!!

IIRC, it was something to do with the system account [It has a seemingly random name] being deleted or corrupt.

If you go into RMMC, can you open the general registry settings or user properties for a specific account?

Will check this out, did you find that all these issues resolved themselves once this was done?

With the slow log ins and slow mapped drive delivery, I have checked switches and they seem okay, yesterday I isolated the servers from the core switch to an edge switch and patched in a laptop, the build still failed?

[andyturpie]

LOG A CALL WITH RM!!

They are there to help you, log it, and escalate it if you need to.

Hi Rad, already had RM in for a remote session, they split my call up into so many segments, with the issues vaguely overlapping and we seem to be heading in circles.

[X-13]

Will check this out, did you find that all these issues resolved themselves once this was done?

With the slow log ins and slow mapped drive delivery, I have checked switches and they seem okay, yesterday I isolated the servers from the core switch to an edge switch and patched in a laptop, the build still failed?

Yup. Almost instantly.

IIRC that's the account used to install software.

[Danp]

Sounds a lot like the issues we had last September, that turned out to be a borked core switch. Replaced that and since then builds have been fine.

Are you fully up to date with updates? Have a look at DWN2970239 and TEC1622957

[Oaktech]

We've had similar issues with our CC3 network. that turned out (as it so often is) to be a conspiracy of problems... a corrupt system account, inconsistent permissions on build areas and one of two trunked fibre lines between our core switch and the rest of the network that had a nicked jacket and was unstable.

[rad]

My bad and apologies,

[andyturpie]

My bad and apologies,

He he no need for apologies

[Sylv3r]

Cannot build machines keeps failing at (installing additional packages)

We had this exact issue just before christmas but none of the other issues you mention - ours had something to do with a corrupt password on the system account connected to the DB. RM 3rd line support dialled in and resolved the issue after a couple of days tinkering and going round in circles with the 1st line support team.

[d-taylor]

We have a batch file that stops RM Service host, clears out the cache files, then starts it again. We have to run this everytime we get a powercut, and also with the odd quirk when it happens. Won't fix your issue, sounds like something much nastier, but RM Service Host isn't very resilient and I resent that I have to run this batch file each time when a vanilla network would just work. I'm sure 3rd line will get to the bottom of your issue.

We have multiple DC's but if Service Host fails network is knackered. There is still no redundancy in CC4 which is just silly.


Something we've also noticed is the RM Database grows over time and RMMC starts to slow (several updates addressed this but it still happens.) RM will dial in and tidy the database, this does help for a couple of years then it happens again.

[andyturpie]

Hi Guys,

Just like to say that RM have been absolutely brilliant and have managed to get to the bottom of the issue which was rectified in a timely and professional manner. I know that there are purists on here which don't hold RM or products in good light, but I think credit where credit is due!

[d-taylor]

Below is the batch file for the FRDC on Server 2008 R2. Running this can solve many CC4 oddities that crop up:
@echo off
rem ********************************
rem Restarts service host and deletes cache
rem **********************************
@echo on

net stop /yes "RM service Host" @echo Waiting for RM service Host To Stop
:loop6 @Rem Wait for 30 seconds
@PING 1.1.1.1 -n 1 -w 30000 >NUL @echo Checking Service State
@sc query "RM service Host" | @FIND "1 STOPPED"
@If %ERRORLEVEL% NEQ 0 goto loop6 @echo Confirmed: RM service Host Service Stopped
@echo Clearing service host cache @Rem Wait for 30 seconds
RD /Q /S "C:\Program Files (x86)\RM\Connect\Comms\HostedProxies"
RD /Q /S "C:\Program Files (x86)\RM\Connect\Comms\RemotingHost"
del /Q /F "C:\ProgramData\RM\Connect\Comms\*.rmcache"
del /Q /F "C:\ProgramData\RM\Connect\Comms\*.cache"
del /Q /F "C:\Users\All Users\Application Data\RM\Networks\RMMC\CacheManager\*.rmcache"

net start /yes "RM service Host" @echo Waiting for RM service Host To Start
:loop7 @Rem Wait for 30 seconds
@PING 1.1.1.1 -n 1 -w 30000 >NUL @echo Checking Service State
@sc query "RM service Host" | @FIND "4 RUNNING"
@If %ERRORLEVEL% NEQ 0 goto loop7 @echo Confirmed: RM service Host Started