Plastic Service Won't Start Error 1053

[rodgeralley]

I'm running into an issue where I can't get the Plastic SCM service to start on one of our Plastic SCM instances.

I get the error:

Windows could not start the Plastic Server 6 service on Local Computer.
Error 1053: The service did not respond to the start or control request in a timely fashion.

Troubleshooting steps I've taken:

• I've rebooted the machine
• I've reinstalled the client
• I've reloaded a the license
• I've tried to swap out the info.dat files in the "\jet\repositories" as you've posted under one of your previous posts.
• I've run "plasticd.exe --server" commands and I get the following:
  • "The BlobData File size cannot be changed on existing storage"
• I've even created the registry key for ServicePipeTimeout

Here's some of the recent info from the debug log:

2018-10-24 15:04:58,498 Main-1    INFO  Daemon - Initialize Datalayer
2018-10-24 15:04:58,576 Main-1    INFO  DatalayerFs.FsConfParser - JET SETTINGS:
2018-10-24 15:04:58,576 Main-1    INFO  DatalayerFs.FsConfParser - basepath=C:\Program Files\PlasticSCM5\server\jet
2018-10-24 15:04:58,576 Main-1    INFO  DatalayerFs.FsConfParser - prefix=
2018-10-24 15:04:58,576 Main-1    INFO  DatalayerFs.FsConfParser - suffix=
2018-10-24 15:04:58,576 Main-1    INFO  DatalayerFs.FsConfParser - maxcachedtrees=50
2018-10-24 15:04:58,576 Main-1    INFO  DatalayerFs.FsConfParser - datafilesize=10737418240
2018-10-24 15:04:58,576 Main-1    INFO  DatalayerFs.FsConfParser - highperf=True
2018-10-24 15:04:58,576 Main-1    INFO  DatalayerFs.FsConfParser - cleaningtemppath=C:\Windows\TEMP\
2018-10-24 15:04:58,623 Main-1    FATAL Daemon - The BlobData file size cannot be changed on existing storage
2018-10-24 15:04:58,623 Main-1    DEBUG Daemon - The BlobData file size cannot be changed on existing storage
   at DatalayerFs.StorageFiles.ValidateBlobDataSize()
   at DatalayerFs.FsDatalayer.Initialize()
   at DatalayerFs.FsDatalayer.Build(FsConf config)
   at Codice.CM.Data.Datalayer.Initialize()
   at akn.a(aas A_0)
   at akn.a(String[] A_0)
2018-10-24 15:05:36,805 Main-1    INFO  Daemon - Initialize Datalayer
2018-10-24 15:05:36,883 Main-1    INFO  DatalayerFs.FsConfParser - JET SETTINGS:
2018-10-24 15:05:36,883 Main-1    INFO  DatalayerFs.FsConfParser - basepath=C:\Program Files\PlasticSCM5\server\jet
2018-10-24 15:05:36,883 Main-1    INFO  DatalayerFs.FsConfParser - prefix=
2018-10-24 15:05:36,883 Main-1    INFO  DatalayerFs.FsConfParser - suffix=
2018-10-24 15:05:36,883 Main-1    INFO  DatalayerFs.FsConfParser - maxcachedtrees=50
2018-10-24 15:05:36,883 Main-1    INFO  DatalayerFs.FsConfParser - datafilesize=10737418240
2018-10-24 15:05:36,883 Main-1    INFO  DatalayerFs.FsConfParser - highperf=True
2018-10-24 15:05:36,883 Main-1    INFO  DatalayerFs.FsConfParser - cleaningtemppath=C:\Windows\TEMP\
2018-10-24 15:05:36,930 Main-1    FATAL Daemon - The BlobData file size cannot be changed on existing storage
2018-10-24 15:05:36,930 Main-1    DEBUG Daemon - The BlobData file size cannot be changed on existing storage
   at DatalayerFs.StorageFiles.ValidateBlobDataSize()
   at DatalayerFs.FsDatalayer.Initialize()
   at DatalayerFs.FsDatalayer.Build(FsConf config)
   at Codice.CM.Data.Datalayer.Initialize()
   at akn.a(aas A_0)
   at akn.a(String[] A_0)

We were having issues with our employees being able to check in large files.  I think something may have accidentally been changed, but I can't figure out what it is.  Let me know what you think I need to do!

[calbzam]

Hi,

Quote

I've tried to swap out the info.dat files in the "\jet\repositories" as you've posted under one of your previous posts

- We have to be careful when editing the Jet metadata files. Could you restore all the Jet modified files, restart the Plastic server and attach the full "plastic.debug.log.txt" file to support@codicesoftwre.com? Please also attach if you have a custom "jet.conf" file.

- Is this your central production server or a new server you are setting up?

Regards,

Carlos.

[calbzam]

According to the error message 

The BlobData file size cannot be changed on existing storage

You have changed the "Max size of blob files" parameter in your server configuration (probably via webadmin). This parameter cannot be modified once you have created databases (and blobs).

From your log:

2018-10-24 15:04:58,576 Main-1    INFO  DatalayerFs.FsConfParser - datafilesize=10737418240  --> 10GB

In my test local server with the default value for this parameter:

2018-10-25 16:57:13,355 Main-1    INFO  DatalayerFs.FsConfParser - datafilesize=4294967296 --> 4GB

 

In order to switch back to 4GB, you can edit the following value at "jet.conf":

...

datafilesize=4

 

Regards,

Carlos.

[rodgeralley]

It was the Blob Datafilesize!  I had set it to 10 in the webadmin!  Okay, so we were having issues with checkins with large file sizes.  I was thinking that the blob files max size was what was causing the issue, which is why I made those changes.  Thank you for helping me fix it!  Do you know what I might need to change in order to allow larger checkins?

[rodgeralley]

The checkins/commits are easily 200MB+

[calbzam]

Regarding the larger checkin, please send us the full debug log to support@codicesoftware.com (and the estimated time you performed the checkin). We will debug the issue.

Carlos.

[manu]

2 minutes ago, rodgeralley said:

The checkins/commits are easily 200MB+

That should be preeeeeetty easy for Plastic so let's check the logs!

How good is the HD used by the Plastic SCM server? Can you run the following test at the Plastic SCM server machine?

cm iostats --disktest --disktestpath=c:\ --diskdatasize=100

We are interested in the "Disk write speed" value, repart the command a couple of times to ensure we get coherent results. Share with us the result

 

 

[rodgeralley]

[rodgeralley]

2018-10-25 14:41:16,287 NetworkThread-21    DEBUG PlasticProto.ConnectionFromClient - conn   46. ReceiveAsync
2018-10-25 14:41:16,287 W-3 00000000-0000-0000-0000-000000000000 skrone DISTI-PC-1223 DEBUG datalayerfs -   File 'C:\Program Files\PlasticSCM5\server\jet\rep_30\branches\names.dat' opened for reading. Size: 3703 bytes.
2018-10-25 14:41:16,287 W-3 00000000-0000-0000-0000-000000000000 skrone DISTI-PC-1223 DEBUG datalayerfs -   File 'C:\Program Files\PlasticSCM5\server\jet\rep_30\blobs\blobs.dat' opened for reading. Size: 4292924861 bytes.
2018-10-25 14:41:16,287 W-3 00000000-0000-0000-0000-000000000000 skrone DISTI-PC-1223 DEBUG datalayerfs - 1 comments resolved in 0 ms
2018-10-25 14:41:16,287 W-3 00000000-0000-0000-0000-000000000000 skrone DISTI-PC-1223 DEBUG datalayerfs -   File 'C:\Program Files\PlasticSCM5\server\jet\rep_30\branches\names.dat' opened for reading. Size: 3703 bytes.
2018-10-25 14:41:16,287 W-3 00000000-0000-0000-0000-000000000000 skrone DISTI-PC-1223 DEBUG datalayerfs - 1 comments resolved in 0 ms
2018-10-25 14:41:16,287 W-3 00000000-0000-0000-0000-000000000000 skrone DISTI-PC-1223 DEBUG Transaction - Commiting implicit transaction
2018-10-25 14:41:16,287 W-3    INFO  ChannelCall - conn:    46 protocol:plasticproto recb:   476|rect: 62|sentb:   340|sendt:  0|queuedt:       0|prt:       0|th:    3|dest:   0|mt:       0|sert:   0|zip:   0|cpu:       0|     10.10.8.255|user::skrone|GetBranchInfoByName
2018-10-25 14:41:16,287 W-3    DEBUG WorkerThread - WorkerThread.Run: Going to GetWork. Id: 3
2018-10-25 14:41:46,319 A-24    DEBUG PlasticProto.ConnectionFromClient - conn   45. awaked. total 1
2018-10-25 14:41:46,319 A-24    DEBUG PlasticProto.ConnectionFromClient - conn   45. Closed.
2018-10-25 14:41:46,319 A-24    DEBUG PlasticProto.ConnectionFromClient - conn   45 from 10.10.8.255 closed
2018-10-25 14:41:46,319 A-23    DEBUG PlasticProto.ConnectionFromClient - conn   46. awaked. total 2
2018-10-25 14:41:46,319 A-23    DEBUG PlasticProto.ConnectionFromClient - conn   46. Closed.
2018-10-25 14:41:46,319 A-23    DEBUG PlasticProto.ConnectionFromClient - conn   46 from 10.10.8.255 closed
2018-10-25 14:42:16,272 W-26    DEBUG WorkerThreadPool - Thread leaving. 5 still running
2018-10-25 14:42:16,287 W-3    DEBUG WorkerThreadPool - Thread leaving. 4 still running
2018-10-25 14:42:40,772 ActivityScheduler-16    DEBUG Security - Start refreshing the MemberInfo cache as scheduled activity
2018-10-25 14:42:40,772 ActivityScheduler-16    INFO  e - ReadGroups. [C:\Program Files\PlasticSCM5\server\groups.conf] loaded in 0 ms. Recursion checked in 0 ms
2018-10-25 14:42:40,772 ActivityScheduler-16    INFO  Security - Reload reloaded 4 users in 0 ms
2018-10-25 14:42:40,772 ActivityScheduler-16    DEBUG Security - Finished refreshing the MemberInfo cache in 0 ms

 

[rodgeralley]

Does that help?

[manu]

14 hours ago, rodgeralley said:

I'm afraid that's super slow for a server machine, Jet will notice the lack of speed

The average speed for a 7200 RPM drive it ranges from about 80-160 MB/s.

The average speed for a regular SSD drive is about 120-300 MB/s

I'll check the logs now.

[manu]

@rodgeralley I think it would be better if you could contact us to support at codicesfotware dot com sharing the full "plastic.debug.log.txt" server log file. We will check for the slowest operation where the problem is, it might be the Server HD speed but we will double check it with the logs.

Is it possible for you to replace the HD? If it's not doable we also offer the Plastic SCM cloud, you can give it a try for free to check if the speed is better. Cloud repositories are always private and can be encrypted if needed.

[rodgeralley]

Sent.  Also, for the record, and I don't know if I need to say it on here or not, but every time I log into this forum, I have to reset my password.  It never saves it.  I can't figure out why.


 

[rodgeralley]

Additionally, for the sake of the Cloud version, we actually looked into it already.  However, cloud services are prohibited by most of our customers for security reasons, so it has to be an on-site service that we offer.  We set this particular one up to be shared with our customers in France (we are located in Orlando, Florida).

And as for the HD, it's a VM sitting on our DMZ.  The same host, also hosts our external services including our company email, without issue.  So i don't know that it would be the hard drive speed, as the VMHost is quite powerful.

[manu]

46 minutes ago, rodgeralley said:

Sent.  Also, for the record, and I don't know if I need to say it on here or not, but every time I log into this forum, I have to reset my password.  It never saves it.  I can't figure out why.


 

Let me report this to the forum support!

[calbzam]

Regarding the password reset issue, this the information from the forum support:

Quote

The issue here was that the standard login was set only to allow display name for login, where the converter one was set to this and email. This meant that email shows on the form. Anyone trying to log in using their email address that wasnt part of your conversion would have therefore failed. When resetting, it logs them in while they are doing so. I have switched them both to accept either display or email, which will correct the problem

 

Regards,

Carlos.

[rodgeralley]

Fixed it!  Thank you!

[manu]

On 10/26/2018 at 3:34 PM, rodgeralley said:

Sent.  Also, for the record, and I don't know if I need to say it on here or not, but every time I log into this forum, I have to reset my password.  It never saves it.  I can't figure out why.


 

I spoke with the Forum support and the issue has been resolved, can you confirm?

Thank you again for reporting the problem.

Best,
Manu.

[rodgeralley]

Confirmed.  It works now.  Thank you!