New builds using Passenger 6.0.8 fail to start up because of file permission issues. Passenger is being loaded as a module in NGINX.
2021/04/06 01:41:46 [emerg] 1#1: mkdir() "/usr/share/nginx/passenger_temp" failed (13: Permission denied)
Same configuration works fine in 6.0.7. What parameter needs to be updated for the path above?
Related
I am trying in vain to Gluu Server in docker container on my macbook . Here are the instructions how to start Gluu Server Single Host . It fails with following error
pmac:single-host root# ./run_all.sh
[I] Determining OS Type and Attempting to Gather External IP Address
Host is detected as Mac
Is this the correct external IP Address: 192.168.1.103 [Y/n]? Y
[I] Preparing cluster-wide config and secret
[I] Checking existing config in Consul
[W] Unable to get config in Consul; retrying ...
[W] Unable to get config in Consul; retrying ...
[W] Unable to get config in Consul; retrying ...
[W] Configuration not found in Consul
docker: Error response from daemon: OCI runtime create failed: container_linux.go:346:
starting container process caused "process_linux.go:449:
container init caused \"rootfs_linux.go:58: mounting \\\"/Users/peeyush/projects/community-edition-containers-4.0/examples/single-host/generate.json\\\"
to rootfs \\\"/var/lib/docker/overlay2/b7c623d9c7347ee09d7bda4dee929b5eb3bbbceb9ff956843cd693fb7fa3cbf5/merged\\\"
at \\\"/var/lib/docker/overlay2/b7c623d9c7347ee09d7bda4dee929b5eb3bbbceb9ff956843cd693fb7fa3cbf5/merged/app/db/generate.json\\\"
caused \\\"open /var/lib/docker/overlay2/b7c623d9c7347ee09d7bda4dee929b5eb3bbbceb9ff956843cd693fb7fa3cbf5/merged/app/db/generate.json:
permission denied\\\"\"": unknown.
Try to run using sudo ./run_all.sh and let me know
For anyone else that arrives at this question using RHEL or Centos, I had the same issue with RHEL 7.3
Apparently In older RHEL kernel versions they deny creation of mount namespaces inside a user namspace because of an out-of-tree patch. apparently RHEL 7.5 will fix this.
I updated my OS to 7.9 and it resolved the issue:
Check for updates:
yum check-update
Update the system
yum update
And then reboot
reboot
After this follow the instructions as documented running as sudo.
I am trying to run XAMPP on my macbook OSX. MySQL and Apache run without a hitch. The ProFTPD server isn't running. I've tried restarting. That didn't work. I went into the xampp files and changed the port (from 21 to 12). I'm not sure what's wrong but I've been receiving the following error message:
Starting ProFTPD...
Exit code: 8
Stdout:
Checking syntax of configuration file
proftpd config test fails, aborting
Stderr:
MacBook-Pro.local proftpd[62092]: warning: unable to determine IP address of 'MacBook-Pro.local'
MacBook-Pro.local proftpd[62092]: error: no valid servers configured
MacBook-Pro.local proftpd[62092]: Fatal: error processing configuration file '/Applications/XAMPP/xamppfiles/etc/proftpd.conf'
Can anyone offer any help? It would be very appreciated. Thank you!
When I attempt to run minikube on locally, I got this error:
minikube start
Starting local Kubernetes v1.8.0 cluster...
Starting VM...
Getting VM IP address...
Moving files into cluster...
E1229 09:52:55.432296 8935 start.go:223] Error updating cluster: Error updating localkube from uri: Error creating localkube asset from url: Error opening file asset: /Users/user007/.minikube/cache/localkube/localkube-v1.8.0: open /Users/user007/.minikube/cache/localkube/localkube-v1.8.0: no such file or directory
How can I fix this problem? I tried many ways, but no one worked.😢
My system is macOS Sierra.
Looks like this is a recurring issue. Check out the github issues here: https://github.com/kubernetes/minikube/issues/2468
I was having this issue and followed Chris Darby link (https://github.com/kubernetes/minikube/issues/2468).
I managed to fix it with the following workaround (considering the new version 1.9.0):
Download localkube
Rename it to localkube-v1.9.0
Copy the file
in ~/.minikube/cache/localkube/localkube-v1.9.0
while executing the below cookbook to configure a site in IIS,
chef-client --local-mode --runlist 'recipe[cookbooksample::web]'
I am getting the error
"cannot create a file when that file already exists" Ran
C:\windows\system32\appcmd.exe start sitename:"testapp" and return
183.
I verified no such file exists and even tried by changing different application pool and website name.
where it went wrong?
I have installed GitLab in a Ubuntu14.04 server with LAMP. I am aware that GitLab has been designed to work with nginx based servers, but this server has web apps that have been installed over apache2 and I need them.
I tried installing GitLab following this tutorial. I used the gitlab_7.7.1-omnibus.5.4.1.ci-1_amd64.deb package.
I ended up having the same problem than some of the people who commented (error 502). I tried following this link through the comments to see if I could find any relevant information, but it doesn't seem to be relevant for my case.
I keep searching for alternative ways to configure nginx through apache2, but I'm not finding anything.
Here's a log entry that I found in /var/log/gitlab/nginx/gitlab_error.log:
2015/01/30 19:32:27 [error] 995#0: *3 connect() to unix:/var/opt/gitlab
/gitlab-rails/sockets/gitlab.socket failed (111: Connection refused)
while connecting to upstream, client: 127.0.0.1, server: mygitlab.com,
request: "GET / HTTP/1.1", upstream: "http://unix:/var/opt/gitlab/gitlab-
rails/sockets/gitlab.socket:/", host: "localhost:8080"
It seems that it is a problem with upstream socket's configuration.
gitlab-ctl status outputs:
run: logrotate: (pid 4465) 1488s; run: log: (pid 826) 5087s
run: nginx: (pid 837) 5087s; run: log: (pid 825) 5087s
run: postgresql: (pid 827) 5087s; run: log: (pid 817) 5087s
run: redis: (pid 828) 5087s; run: log: (pid 818) 5087s
run: sidekiq: (pid 839) 5087s; run: log: (pid 824) 5087s
run: unicorn: (pid 5861) 0s; run: log: (pid 823) 5087s
Anyways, it seems that Omnibus doesn't install nginx at all (or I guess so, there's no service named 'nginx' in /etc/init.d/. I do not discard the possibility that the log generator confused apache2 with nginx).
Questions:
Is it mandatory to install nginx to run gitlab? If so, can it co-exist with apache properly?
Is proxying GitLab through apache a good approach to what I'm looking for?
I don't know much about Unicorn. I found this question in SO. Does this mean that Unicorn handles upstream sockets in GitLab?
EDIT: Checking this document I decided to check if Omnibus installed Ruby with ruby -v and I noticed that Ruby wasn't installed at all. So I decided to compile and install it, following the document's instructions:
mkdir /tmp/ruby && cd /tmp/ruby
curl -L --progress http://cache.ruby-lang.org/pub/ruby/2.1/ruby-2.1.5.tar.gz | tar xz
cd ruby-2.1.5
./configure --disable-install-rdoc
make
sudo make install
Anyways, it seems some dependencies failed and ruby is not plently operative. When I try sudo gem install bundler --no-ri --no-rdoc, the following errors ocurr:
ERROR: Loading command: install (LoadError)
cannot load such file -- zlib
ERROR: While executing gem ... (NoMethodError)
undefined method `invoke_with_build_args' for nil:NilClass
During the make operation, these errors happened:
Failed to configure -test-/win32/dln. It will not be installed.
Failed to configure -test-/win32/dln/empty. It will not be installed.
Failed to configure -test-/win32/fd_setsize. It will not be installed.
Failed to configure dbm. It will not be installed.
Failed to configure fiddle. It will not be installed.
Failed to configure gdbm. It will not be installed.
Failed to configure openssl. It will not be installed.
Failed to configure readline. It will not be installed.
Failed to configure tk. It will not be installed.
Failed to configure tk/tkutil. It will not be installed.
Failed to configure win32. It will not be installed.
Failed to configure win32ole. It will not be installed.
Failed to configure zlib. It will not be installed.
EDIT2: It seems that compiling and installing Ruby was completely inneccessary. Looking around for config files, I saw that Omnibus installed lots of dependencies (Ruby among them) in /opt/gitlab/embedded/ directory. :-S
Is it mandatory to install nginx to run gitlab? If so, can it co-exist
with apache properly?
Gitlab is bundled with nginx but you can use it with Apache HTTP server.
Is proxying GitLab through apache a good approach to what I'm looking
for?
You can disable nginx in Gitlab config and configure Apache to forward to Gitlab.
I don't know much about Unicorn. I found this question in SO. Does
this mean that Unicorn handles upstream sockets in GitLab?
Unicorn is HTTP server for Ruby. You can forward requests to Gitlab Unicorn server or gitlab-workhorse.
Below is the basic /etc/gitlab/gitlab.rb configuration used with Apache, other lines are commented:
external_url 'https://my.example.com/gitlab'
gitlab_workhorse['enable'] = true
gitlab_workhorse['listen_network'] = "tcp"
gitlab_workhorse['listen_addr'] = "127.0.0.1:9191"
unicorn['listen'] = '127.0.0.1'
unicorn['port'] = 9099
web_server['external_users'] = ['www-data']
web_server['username'] = 'apache' #'gitlab-www'
web_server['group'] = 'apache' #'gitlab-www'
nginx['enable'] = false
And make proxyPass in /etc/apache2/apache2.conf to the gitlab-workhorse running on 9191:
ProxyPass /gitlab http://127.0.0.1:9191/gitlab
RequestHeader add X-Forwarded-Proto "https"
RequestHeader set X-Forwarded-Ssl on