AWS Thinkbox Discussion Forums

deadline mobile error

Trying to set up deadline mobile. It seems to make a connection to pulse, but then i get this crash in pulse:

An unhandled exception occurred: Method not found: ‘Mono.Security.Protocol.Tls.SslServerStream…ctor’. (System.MissingMethodException)
at System.Net.EndPointListener.OnAccept (System.Object sender, System.EventArgs e) [0x00000] in :0
at System.Net.Sockets.SocketAsyncEventArgs.OnCompleted (System.Net.Sockets.SocketAsyncEventArgs e) [0x00000] in :0
at System.Net.Sockets.SocketAsyncEventArgs.AcceptCallback (IAsyncResult ares) [0x00000] in :0
at System.Net.Sockets.SocketAsyncEventArgs.DispatcherCB (IAsyncResult ares) [0x00000] in :0

After which the mobile phone app also pops an error (disappears too fast)

Hmm, it sounds like there might be a missing Mono package on the machine. Which Linux distro are you using? If the version you have lets you install Mono from a package manager, check to see if you have the mono-complete package installed.

Its centOS 6.3

Doesnt seem to have mono installed, i thought deadline installer adds those packages? The package manager doesnt show it neither

the yum install locations for mono still point to the now defunct novell site…

The Deadline installer doesn’t handle the Mono installation on Linux, and the mono packages available in CentOS’s package manager are very out of date. We do have a handy shell script that can be used to automatically download, build, and install Mono 2.10.9 on a system though. I’ve attached it here.

It installs all the Mono components and puts everything in /opt/mono-2.10.9. The script adds this to the PATH environment variable as the first path, so it should take precedence over any existing Mono installation. Just note that you’ll need to log off and log back in for the PATH change to take effect.

Cheers,

Got to love linux, a simple shell script and voila! you have your own mono build hehehe

This has been the most gloriously simple mono build i have ever done. Mobile works too!

Actually, it worked for a while, then i got out of the app.

Now whenever i try to get in, it just crashes right away, even without a data connection :frowning:

Restarting the phone didnt help, the only way to get it working again was to wipe the app’s data, and setup everything from scratch again.

After that, when i clicked on a job, no information was displayed, and the app popped this error:
An error occured while parsing the data from Pulse: somthing about parsing a document

Then this:
An error occured while connecting to Pulse: socket closed

Then the first error a couple of times. Then after a couple of tries, i got some job info to show!

A bit finicky, but now it seems to work again. (damn… spoke too soon, its not starting again… :\ )

Android v1.1

I think the startup problems happen when i set a custom filter up. I set it to not show complete jobs, it worked for a while, but quitting then reentering the app will make it crash right away.

We get a server connection error on iphones, on androids the job list refreshes just fine though.

Tried both wifi/3g

Actually, some other androids fail as well. Maybe its more like, first connection is served, all others are not?

saw an error like this in the pulse log, maybe that helps:

An unhandled exception occurred: The object was used after being disposed. (System.ObjectDisposedException)
at System.Net.ResponseStream.EndWrite (IAsyncResult ares) [0x00000] in :0
at Deadline.Pulses.PulseWebService.a (IAsyncResult A_0) [0x00000] in :0

Actually, its probably not the filters… i kept them untouched, and the app crashes again for me. Oh well :slight_smile:

Can you open a web browser and enter in this URL:

http://PULSESERVER:8080/Mobile_GetJobList

You just need to replace PULSESERVER with the actual pulse server machine name (or IP address). This is basically what the Android app is doing to get info from pulse. Can you then copy the results into a text file, zip it up, and send it to us? We could then use it to try and reproduce on our end.

Thanks!

  • Ryan

Sent to your email!

Thanks! I tested it here (by having Pulse return the contents of the XML file instead of building it up on the fly), and it worked fine. Are you still having issues with the mobile apps? Do you always get errors or is it intermittent?

Thanks!

  • Ryan

On Saturday Pulse went down after the Web Service was used. Unfortunately I don’t have much more information about the crash. The last entry in the log says “[…] Web Service - Executing Mobile_GetJobDetails script…”. I will send the whole log file by mail.

-Christian

After the problems on Tuesday I turned of the web service. Yesterday the farm went down again. This are the last lines of the log file:

2013-04-17 13:31:41: Clean Up Thread - Performing house cleaning 2013-04-17 13:31:41: Cleaning up orphaned tasks 2013-04-17 13:31:41: Performing Job Repository Scan... 2013-04-17 13:31:41: Loading jobs 2013-04-17 13:31:41: Scanning jobs 2013-04-17 13:31:41: Done. 2013-04-17 13:32:40: Clean Up Thread - Performing house cleaning 2013-04-17 13:32:40: Purging old job and slave reports 2013-04-17 13:32:40: Purging job reports for '516f0304c062930b4809f648' because the job no longer exists 2013-04-17 13:32:40: Purging job reports for '516f052d47f5c60d48f3914a' because the job no longer exists 2013-04-17 13:32:40: Cleaning up orphaned tasks 2013-04-17 13:32:40: Performing Job Repository Scan... 2013-04-17 13:32:40: Loading jobs 2013-04-17 13:32:40: Scanning jobs 2013-04-17 13:32:40: Done. 2013-04-17 13:32:40: Purging unsubmitted jobs 2013-04-17 13:33:40: Clean Up Thread - Performing house cleaning 2013-04-17 13:33:40: Performing Job Repository Scan... 2013-04-17 13:33:40: Loading jobs 2013-04-17 13:33:40: Scanning jobs 2013-04-17 13:33:40: Done. 2013-04-17 13:34:39: Clean Up Thread - Performing house cleaning 2013-04-17 13:34:39: Purging old statistics 2013-04-17 13:34:39: Purging slave statistics that are older than Dec 18/12 13:34:39 2013-04-17 13:34:39: Purging repository statistics that are older than Dec 18/12 13:34:39 2013-04-17 13:34:39: Performing Job Repository Scan... 2013-04-17 13:34:39: Loading jobs 2013-04-17 13:34:40: Scanning jobs 2013-04-17 13:34:40: Done. 2013-04-17 13:35:30: Web Service - Executing Mobile_GetJobList script... 2013-04-17 13:35:32: Server Thread - TIMEOUT (unknown client): received no request 2013-04-17 13:35:39: Clean Up Thread - Performing house cleaning 2013-04-17 13:36:00: Performing Job Repository Scan... 2013-04-17 13:36:00: Loading jobs 2013-04-17 13:36:00: Scanning jobs 2013-04-17 13:36:00: Done. 2013-04-17 13:36:00: Web Service - Done executing Mobile_GetJobList script... 2013-04-17 13:36:00: Web Service - Web Service listening for requests... 2013-04-17 13:36:10: Web Service - Executing Mobile_GetJobDetails script... 2013-04-17 13:36:14: Web Service - Done executing Mobile_GetJobDetails script... 2013-04-17 13:36:14: Web Service - Web Service listening for requests... 2013-04-17 13:36:16: Web Service - Executing Mobile_GetJobDetails script... 2013-04-17 13:36:18: Web Service - Done executing Mobile_GetJobDetails script... 2013-04-17 13:36:18: Web Service - Web Service listening for requests... 2013-04-17 13:36:26: Server Thread - TIMEOUT (unknown client): received no request 2013-04-17 13:36:28: TypeError 2013-04-17 13:36:28: : 2013-04-17 13:36:28: invalid result type from FilterLogBoxModel.data() 2013-04-17 13:36:28: TypeError 2013-04-17 13:36:28: : 2013-04-17 13:36:28: invalid result type from FilterLogBoxModel.data() 2013-04-17 13:36:28: TypeError 2013-04-17 13:36:28: : 2013-04-17 13:36:28: invalid result type from FilterLogBoxModel.data()

The logs from the last days will follow by email.

-Christian

Thanks for the logs. This crash looks like it has something to do with the log window. I’m wondering if we should be doing the same thing with Pulse that we did with the Slaves when we removed the log window from the main interface…

Privacy | Site terms | Cookie preferences