Description of problem: Let's get the ball rolling on this one... http://fedoraproject.org/wiki/Features/SysVtoSystemd Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info:
Created attachment 522389 [details] apgd socket
Created attachment 522390 [details] apgd service
Passes simple telnet test... [root@valhalla system]# systemctl start apgd.socket [root@valhalla system]# systemctl status apgd.socket apgd.socket - Authd Activation Socket Loaded: loaded (/lib/systemd/system/apgd.socket) Active: active (listening) since Fri, 09 Sep 2011 19:10:05 +0000; 3s ago Accepted: 0; Connected: 0 CGroup: name=systemd:/system/apgd.socket [root@valhalla system]# telnet localhost 129 Trying ::1... Connected to localhost. Escape character is '^]'. Hibdeot. (Hib-de-ot-PERIOD) Sickwegs (Sick-wegs) cozunet1 (coz-un-et-ONE) ottIfsUn (ott-Ifs-Un) PretFink (Pret-Fink) RekZuor6 (Rek-Zu-or-SIX) Connection closed by foreign host. And note there is still a (tight) window to have this in F16
Created attachment 522525 [details] Agpd service
Once package and shipped your package should no longer have to depend on xinetd https://fedoraproject.org/wiki/Packaging:Guidelines:Systemd https://fedoraproject.org/wiki/Packaging:ScriptletSnippets#Systemd
If I can recall correctly you ping me about this one not working where you mentioned that this was not working. I'm wondering if you are hitting a bug in systemd .35 where the socket out put is broken for in.inetd daemons You can add StandardOutput=socket to the service file to see if you are encountering that bug. Node that this option is only needed since this only affects .35 ( not 34 or earlier ) and will be fixed in 36
Time is late and my english is starting to suck more then usual basically you should only add StandardOutput=socket to the service to test if you are bit by the .35 bug. No need to add it otherwize since that bug will be fixed in .36 and only affects .35 release
Yeah, it's .35, so that could be the issue. ;) Will give that a try, thanks!
Sorry for the long delay here. Ran into two anoying problems: 1) didn't realize that the service file needed to be apg@.service instead of apg.service. 2) apgd needs /dev/random, so in my testing I kept running into it running out of entropy and hanging. Anyhow, all fixed and commited to rawhide. Thanks.