I am trying to use an RHEL 5 Docker container to run Jenkins jobs, but RHEL 5 does not support anything more recent than Java 8. Is there a way to run a Jenkins job in this container without using the Jenkins agent to communicate to the Jenkins server? Maybe with SSH?
(Don't ask me why RHEL 5, it just is. I can't change it.)
I have a customer who is excited about the idea of Docker containers as Jenkins build nodes, but a few of their pipelines still have dependencies on Java 8. Can a Docker container have Java 11 and Java 8, use 11 to run as an agent but 8 to execute a pipeline?
I am trying to get ClearCase 9.1.0.4 Server running on a Docker container to use the container as a Jenkins build slave. I only install the Server version of ClearCase in the container so it doesn't try to install MVFS, and through device sharing I share the MVFS device on the host.
Everything seems to works fine, I can lsview, lsvob, etc. but if I try to setview I get the following error:
cleartool: Error: Couldn't set view tag <any_view>: unknown error in MVFS file system
The setview command works just fine on the host. No error log is generated on the host or the container to tell me what happened.
What is even stranger is that startview works fine on the container! That command mounts the view on /view and I can cd straight into the vob that I need to work on, but setview still complains about the tag.
Anyone seen this before? Did you fix it? Is it possible to have MVFS tell me something more useful than "unknown error"? How?
And yes, I already read this.
https://www.ibm.com/support/pages/system/files/inline-files/docker_clearcase.pdf
It doesn't help and I can't find the person who wrote it.
I am trying to define a sites-available .conf entry for a mediawiki scratch installation on a Raspberry Pi 4 running Raspbian in my local network (behind the firewall). No DNS or domain name associated with it. No matter how I configure it for Apache 2 it comes back with a 500 error. Does anyone have a simple sample mediawiki.conf file that I can use to define the site? The local IP is 10.0.0.67.
I created a private docker registry using AWS ECR. I want to use it with a Kubernetes cluster managed by Rancher, but the password to the registry is changed by AWS every 12 hours. How do I keep it updated as a secret in the cluster?
I have a Rancher 2.5 cluster running on AWS EKS. My CI is done in CodeBuild via a webhook to a Github repo. The buildspec.yml works fine to run the build, tag the containers and then send them to my private registry on AWS ECR, but I have not found an easy way to then trigger an update to the deployment on the cluster to let it know a new container version is available. How do I do that?
I am using Rancher 2.5.8 to run Kubernetes clusters. In my org, we maintain a private docker registry on AWS ECR. I know I can create registry credentials within a project (for use in one or all projects), but ECR changes the password every 12 hours and Rancher has no built in away to update the credentials. Someone came up with a convoluted way of updating the password with two container apps, but I find it much easier to use the aws cli and kubectl in the host vm to do this. I created a cron job that uses kubectl to delete and create a docker-registry secret in the cluster programmatically (there's no way to update a secret, apparently), but I can't figure out how to make that docker-registry secret visible in Rancher. Anyone know how to do this?
The Terraform docs for some weird reason do not explain what "Error: Cycle" means. I've looked everywhere but there is no mention of it on the official docs. (Turns out it is well-known term, a circular dependency, that someone apparently renamed thinking it would make them sound cool...)
I'm having trouble completing a bind to our LDAP servers on Centos 7.1 servers. Manual bind works, but ldapsearch fails with an error:
sssd_be: GSSAPI Error: Unspecified GSS failure. Minor code may provide more information (Message stream modified)
Kerberos is working fine, the tickets are valid. I can't find any documentation about this error. Any clues?
I have a strange requirement that I thought would be easy to do but turns out to be more complicated than I thought.
I run a heavily firewalled HPC scientific cluster. A user wants a Windows 7 VM to run a specific app. I created the VM on my Mac and then copied it to a Centos 7 server in the cluster. VirtualBox runs it fine as a headless server and says it is listening on port 3389.
To get to the cluster you first have to ssh to a login node, and from there you can get the rest of the cluster.
So, I have to create an SSH tunnel, right? Except the tunnel is not to the login node, it's to another server in the cluster after you get into the node. The VM is not running on the login server (that is a VM in and of itself). So I have to do double-forwarding?
I tried using localforward in the .ssh/config file and it doesn't quite work.
Now I'm thinking that I have to configure the login node to forward all traffic on port 3389 to the server running the vm.
Am I moving in the right direction or barking up the wrong tree?
I have a problem in a server that has several dozen snapshots of a ZFS fileset. Back in November we received files that turned out to be for another organization. The content of the files is very sensitive, so the files were deleted. However, by the time we deleted the files, some 42 snapshots had been created of the fileset that contain data or references to those files.
The problem now is that we have to remove that data from the snapshots but we do not want to lose the rest of the contents of the snapshots. Everything I have read so far about ZFS snapshots is that they are all-or-nothing entities. You can create them and destroy them but not modify them to partially remove contents.
Is that indeed the case, or does anyone know of a solution for this?
I would like to create a cron job that runs chkrootkit, but only emails me the resulting log if there is a positive, false or not. The chkrootkit faq gives this as a suggestion:
0 3 * * * (cd /path/to/chkrootkit; ./chkrootkit 2>&1 | mail -s "chkrootkit output" root)
...but I run an HPC cluster with over 100 nodes. I only want it to mail the log if it has detected something to which I should pay attention.
How should I go about doing this?
I have an AWS VM running Ubuntu with Apache2. I have two vhosts, both pointing to different Wordpress installs. I want to set up a Squid reverse proxy for only one of the vhosts.
My squid.conf reads:
http_port 80 accel defaultsite=wordpress.mysite.com no-vhost
cache_peer wordpress.mybackendserver.com parent 80 0 no-query originserver name=myAccel
acl our_sites dstdomain wordpress.mysite.com
http_access allow our_sites
cache_peer_access myAccel allow our_sites
cache_peer_access myAccel deny all
To test it, I set up an entry for wordpress.mysite.com in /etc/hosts pointing to the ip of the Squid server.
When I try to go to the site with the wordpress.mysite.com URL I get the default Apache2 home page.
Is there something I need to do in the http_port statement to get the url passed correctly to Apache so it can return the content for the right vhost?