Quote:
|
Originally Posted by jre
I just read some things about SIGHUP (didn't know anything about it before). From that it sounds as if you changed the code so that you can 'reload' (as defined in the LSB) the daemon (peerguardnf). So it isn't necessary to restart the daemon when a new blocklist was installed. Am I right with this??? (I hope so because I missed this feature)
|
The source code I downloaded from sourceforge just reacted to SIGTERM (for shutting down the program). I added SIGHUP for log file reopening (which is common among other daemons, as far as I know). So at the moment there is no other reaction to sending SIGHUP than reopening the log file. But you are right about the reloading of configuration information. This is not handled yet - though it would be nice.
Quote:
|
Originally Posted by jre
Never knew there was a CVS version. Now I can tell you that you're definitely not dumb! I've just downloaded, compiled and installed it.
I'd propose you call your package peerguardnf_1.5CVS-[somenumber]_i386.deb to make it clear you used the CVS code. I think most other people always used the 1.5beta (which was 1-2 months older than the last changes in the CVS).
|
Yes. That's true. I'll change that when I'm uploading something new.
Quote:
|
Originally Posted by jre
The file size of /usr/bin/peerguardnf is now 690196 here. Is this normal? 9 times bigger then your version, but still just 2/3 of the binary from the official 1.5beta debian package. Why this big differences?
|
I don't know. I just ran ./configure && make.
Quote:
|
Originally Posted by jre
The changes from 1.5beta to CVS I saw (without looking into the code):
- QUEUE is also put in the FORWARD rule.
- There's a message in the log: 'Deamon forked, pid is 23202' (The typo is not mine  ) |
There's much more in the code (I had a glance at a recursive diff. Too much for this evening
)
For the time being, I uploaded a new package (it's attached to my first post.)
Code:
peerguardnf (1.5b-3) unstable; urgency=low
* updated Main.cpp to create /var/run/peerguardnf.pid instead of
/var/run/pg.pid (better for /etc/init.d/peerguardnf)
* increased timeout from 1 to 2 seconds while checking for list updates
-- clessing <clessing@freenet.de> Thu, 2 Mar 2006 21:35:38 +0100
peerguardnf (1.5b-2) unstable; urgency=low
* moved logfile to /var/log/PG.log
* fixed config to just open port 80 and 443
* changed blocklist update to check for phoenixlabs.org instead of
www.sourceforge.net
* updated NEWS to point to http://forums.phoenixlabs.org/
-- clessing <clessing@freenet.de> Thu, 2 Mar 2006 20:41:44 +0100
peerguardnf (1.5b-1) unstable; urgency=low
* Initial release
* daily rotation of logfile
* daily syncing with bluetack.co.uk
* start/stop scripts in init.d
-- clessing <clessing@freenet.de> Thu, 2 Mar 2006 12:39:30 +0100