SCM Manager returning various error after Migration to v2

SCM-Manager Bug Reports SCM-Manager
1

Good evening,

Today I upgraded scm from version 1.60 to 2.30 only that it returns a 404 error from Google console (the browser just keeps loading until the request times out).

running journalctl -u scm-server returns the following error:

Exception in thread "ScmEventBus-1-0" java.lang.UnsatisfiedLinkError: /tmp/jna6511840250923886855.tmp: /tmp/jna6511840250923886855.tmp: failed to map segment from shared object`
at java.lang.ClassLoader$NativeLibrary.load(Native Method)`
at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1934)
at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1817)
at java.lang.Runtime.load0(Runtime.java:810)
at java.lang.System.load(System.java:1088)
at com.sun.jna.Native.loadNativeDispatchLibraryFromClasspath(Native.java:1018)
at com.sun.jna.Native.loadNativeDispatchLibrary(Native.java:988)
at com.sun.jna.Native.<clinit>(Native.java:195)
at com.sun.akuma.CLibrary.<clinit>(CLibrary.java:89)
at com.sun.akuma.JavaVMArguments.resolvePID(JavaVMArguments.java:128)
at com.sun.akuma.JavaVMArguments.ofLinux(JavaVMArguments.java:116)
at com.sun.akuma.JavaVMArguments.of(JavaVMArguments.java:104)
at com.sun.akuma.JavaVMArguments.current(JavaVMArguments.java:92)
at sonia.scm.lifecycle.PosixRestartStrategy.restart(PosixRestartStrategy.java:57)
at sonia.scm.lifecycle.PosixRestartStrategy.executeRestart(PosixRestartStrategy.java:47)
at sonia.scm.lifecycle.RestartStrategy.restart(RestartStrategy.java:67)
at sonia.scm.lifecycle.BootstrapContextFilter.handleRestartEvent(BootstrapContextFilter.java:99)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:498)
at com.github.legman.InvocationContext.invoke(InvocationContext.java:108)
at com.github.legman.InvocationContext.proceed(InvocationContext.java:101)
at com.github.legman.micrometer.MicrometerInvocationInterceptor.invoke(MicrometerInvocationInterceptor.java:47)
at com.github.legman.InvocationContext.proceed(InvocationContext.java:99)
at com.github.legman.EventHandler.handleEvent(EventHandler.java:103)
at com.github.legman.SynchronizedEventHandler.handleEvent(SynchronizedEventHandler.java:52)
at com.github.legman.EventBus.dispatchSynchronous(EventBus.java:452)
at com.github.legman.EventBus.lambda$dispatch$1(EventBus.java:444)
at org.apache.shiro.subject.support.SubjectRunnable.doRun(SubjectRunnable.java:120)
at org.apache.shiro.subject.support.SubjectRunnable.run(SubjectRunnable.java:108)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)`*

From google chrome console I get the following error:

GET http://serverip/scm/api/v2/ 404 (Not Found) ApiClient.request @ apiclient.ts?718f:307

If I force restart from the server with service scm-server restart , scm manager returns up and running, and with the plugin installed.

Now it seems like the browser is finding it difficult to successfully restart SCM service even if in scm-manager.log I can see the jvm service was restarted.

I also checked the apache log error_log and found an error as follows:

[proxy:error] [pid 32822] (111)Connection refused: AH00957: HTTP: attempt to connect to 127.0.0.1:8080 (localhost) failed
[proxy_http:error] [pid 32822] [client 10.12.173.165:62193] AH01114: HTTP: failed to make connection to backend: localhost
[proxy:error] [pid 32821] (111)Connection refused: AH00957: HTTP: attempt to connect to 127.0.0.1:8080 (localhost) failed

The thing is with version 1.60 I never got any of these errors, the reverse proxy settings in httpd.conf works until this restart from console activity.

I really apologize for the long story, I am only trying to make it comprehensive as much as possible.

Thanks to anyone willing to help a brother.

Regards

2

Hey @igwuborchuks,

we will have a look and come back to you asap.

Regards, Eduard

1 Like
3

I’m not sure yet whether this is related but could you please describe your environment a little more?

Which operating system, which architecture, which Java version do you use? Does this error also occur if you start SCM-Manager 2.30.0 clean (without migration)?

Also: Could you check if the user running SCM-Manager have proper permissions?

4

Hello eheimbuch,

I run a: rhel 8

openjdk version "1.8.0_302"
OpenJDK Runtime Environment (build 1.8.0_302-b08)
OpenJDK 64-Bit Server VM (build 25.302-b08, mixed mode)

I followed your suggestion and installed a clean version of scm 2.30, everything worked fine, installed a plugin and the page refreshed without any problems.

I still get the proxy-error but I think that is related to the httpd reverse proxy configuration which the security team is currently verifying as apache was also recently updated to 2.24

I think the issue might be the process of migrating from version 1.60 to 2.30.
My procedure for the migration from v1.60:

  1. remove the /opt/scm-server directory (v1.60 installation)

  2. I do not touch the /var/lib/scm/ folder (where repos and plugins ecc are stored)

  3. update /etc/yum.repos.d/SCM-Manager.repo:

    [scm-manager]
name=SCM-Manager Repository
baseurl=https://packages.scm-manager.org/repository/yum-v2-releases/
enabled=1
gpgcheck=1
priority=1
gpgkey=file:///etc/pki/rpm-gpg/SCM-Manager

  4. executed:
    curl -o /etc/pki/rpm-gpg/SCM-Manager https://packages.scm-manager.org/repository/keys/gpg/oss-cloudogu-com.pub

  5. dnf install scm-server : which installs scm-server 2.30.1 and mercurial 4.8.2-1

I started scm-server service and in the migration window, selected the move method, the window starts the migration but then does not refresh the page.

Am I missing a process in the migration flow?.

Thank you once again.

Regards

5

Thanks for the details. We assume that our restart mechanism somehow doesn’t work on your machine. Could you try to simply restart the server manually?

6

Manually restarting the scm service works.

However, I did a fresh installation of the v1.60 and then an upgrade to 2.30 and surprisingly the service did restart via web (during migration and even during plugin installation) .

At this point it works on a fresh installation of 1.60… I observed that the permission after the migration to 2.30 for /var/lib/scm changed to drwxr-xr-x rather than the drwx------ for version 1.
Is there a standard permission needed for /scm directory??.. I mean a 775 for everything in the /var/lib/scm directory.

Thank you once again.

7

Hi,

I fixed it.
After installation, I just ran chmod -R 775 scm and scm was able to restart service from browser

1 Like
8

I have closed this thread since there was no activity in the last 30 days.