| 1 |
On Tue, February 18, 2014 10:47, Alan McKinnon wrote: |
| 2 |
> On 18/02/2014 05:46, Mark David Dumlao wrote: |
| 3 |
>> I used to use cherokee. Fast, light, awesome, and with a web admin. |
| 4 |
>> The init script always failed me. /etc/init.d/cherokee stop was not a |
| 5 |
>> guaranteed stop to all forked cherokee processes - the parent pid |
| 6 |
>> dies, but some forked process or something, usually related to |
| 7 |
>> rrdtool, doesn't. Or the parent does exit and erases the pid file but |
| 8 |
>> it returns control immediately and its not yet done exiting. Something |
| 9 |
>> like that or other. Point is, I've several times had to ps aux|grep |
| 10 |
>> ... kill; zap; start - on production servers. |
| 11 |
> |
| 12 |
> |
| 13 |
> Valid point. Other than vixie-cron (damn thing just never seems to die |
| 14 |
> properly on any platform so restarts always fail) I don't really run |
| 15 |
> into these issues |
| 16 |
|
| 17 |
Interesting, I have never had issues with restarting vixie-cron using the |
| 18 |
supplied init-scripts. |
| 19 |
|
| 20 |
> What I do run into is daemons that drop privs on start up, like |
| 21 |
> tac_plus. Unwary new sysadmins always try start/stop it as root, causing |
| 22 |
> an unholy mess. Root the owns the log and pid files, when tac_plus drops |
| 23 |
> privs it can't record it's state so continues to service requests but |
| 24 |
> fails to log any of them. For an auth daemon, that's a serious issue. |
| 25 |
|
| 26 |
Shouldn't sysadmins use the init-scripts for that? |
| 27 |
If done correctly, permissions should not be an issue. |
| 28 |
|
| 29 |
Restarting services without keeping file ownership into account will |
| 30 |
always cause issues. Regardless of the init-system used. |
| 31 |
|
| 32 |
And tac_plus not checking if it is allowed to write to the log during the |
| 33 |
initialization phase should be considered a bug. |
| 34 |
|
| 35 |
-- |
| 36 |
Joost |
Archives