Mongrel.pid log does not exist

Rails Deployment
1

I’m getting the following error when I stop mongrel… I know its
running but I need the log files. It appears the log isn’t being
created / written to. I this message:

!!! PID file log/mongrel.8000.pid does not exist. Not running?

This is my mongrel_cluster.yml

cwd: /var/rails/soca/current
port: “8000”
environment: production
address: 127.0.0.1
pid_file: log/mongrel.pid
servers: 2

…and I’ve the following permissions on the directory:

drwxrwxrwx 2 bleroux root 4096 Feb 19 16:44 log

Which is here:

[root@server1 current]# pwd
/var/rails/soca/current

Any ideas??

2

Brian,

Have you tried using a full path for: pid_file: ?

–
Aníbal Rojas

3

On 2/20/07, Brian LeRoux [email protected] wrote:

cwd: /var/rails/soca/current

Which is here:

[root@server1 current]# pwd
/var/rails/soca/current

Any ideas??

It is a known bug. You should specify full path log and pid files for
mongrel_cluster.

–
Luis L.
Multimedia systems

Leaders are made, they are not born. They are made by hard effort,
which is the price which all of us must pay to achieve any goal that
is worthwhile.
Vince Lombardi

4

This may be a silly question, but does the pid file actually exist?
Have you checked yourself?

I’m running into an issue like this myself, which seems to be
triggered by a slow shutdown of one thread (on port 8001, say).
Capistrano doesn’t wait for the shutdown and tries to do a restart,
but when the new thread on 8001 starts it can’t write the pid file
because the old one is still shutting down. Then the next time I do a
rake deploy, the pid file doesn’t exist and I lose touch with the
thread.

At least, that’s my best current theory.

5

Ok —thanks guys, I tried specifying a full path for the pid but still
no go.
This must be a permissions issue… but the directory is open to the
world to modify. This is odd.

6

I finally figured it was permissions issues the whole time… which is
my dumb. =) Runs like a champ now!

7

I am having this exact same issue, and perms for log/ seem correct:

kmarsh@host:shared$ ll
total 20K
drwxr-xr-x 3 kmarsh 4.0K Feb 15 01:36 assets/
drwxrwxrwx 2 kmarsh 4.0K Feb 23 14:36 log/
drwxr-xr-x 595 www 8.0K Feb 23 18:22 photos/
drwxrwxr-x 2 kmarsh 4.0K Dec 11 16:06 system/

Any ideas?

8

Can I see your mongrel config?

9

Brian you mention it being a known bug, is the fact that sometimes
when restarting mongrel clusters I will get an error stating that the
PID already exists?

Almost like the restart process didn’t give the mongrels enough time
to shutdown before attempting to start them up again.