[Bug 594223] New: priorititize sshd on boot
http://bugzilla.novell.com/show_bug.cgi?id=594223 http://bugzilla.novell.com/show_bug.cgi?id=594223#c0 Summary: priorititize sshd on boot Classification: openSUSE Product: openSUSE 11.3 Version: Factory Platform: All OS/Version: Linux Status: NEW Severity: Enhancement Priority: P5 - None Component: Basesystem AssignedTo: bnc-team-screening@forge.provo.novell.com ReportedBy: jengelh@medozas.de QAContact: qa@suse.de Found By: Beta-Customer Blocker: --- There should be some way to start sshd immediately after network, and network ASAP. Right now, my rc3.d is a horrible mess with slow-starting services coming first S01acpid S01dbus S01earlysyslog S01microcode.ctl S01random S02haldaemon S02network S03syslog S04nmb S04rpcbind S05nfs S06kbd S07clamd S07irq_balancer S07ldap S07lm_sensors S07mysql S07network-remotefs S07saslauthd S07spamd S07sshd S07xinetd .. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
http://bugzilla.novell.com/show_bug.cgi?id=594223
http://bugzilla.novell.com/show_bug.cgi?id=594223#c
yang xiaoyu
http://bugzilla.novell.com/show_bug.cgi?id=594223
http://bugzilla.novell.com/show_bug.cgi?id=594223#c1
Anna Bernathova
http://bugzilla.novell.com/show_bug.cgi?id=594223
http://bugzilla.novell.com/show_bug.cgi?id=594223#c2
Dr. Werner Fink
http://bugzilla.novell.com/show_bug.cgi?id=594223
http://bugzilla.novell.com/show_bug.cgi?id=594223#c3
Jan Engelhardt
http://bugzilla.novell.com/show_bug.cgi?id=594223
http://bugzilla.novell.com/show_bug.cgi?id=594223#c4
Dr. Werner Fink
BTW, I do not believe anybody is really mounting /usr from the network, or that this scenario has been tested, because there are quite a handful of services started before nfs.
It is a matter of fact that this is done e.g. for large clusters of small systems sharing *one* /usr parittion. And we had indeed bugreports that this feature was broken in past. For you problem: simply replace kbd with clamd and see what happens. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
http://bugzilla.novell.com/show_bug.cgi?id=594223
http://bugzilla.novell.com/show_bug.cgi?id=594223#c5
Jan Engelhardt
for large clusters of small systems sharing *one* /usr parittion
How well does that even integrate with rpm? Updating a client would mean either attempting to write to a read-only /usr, or overwriting the files in /usr, thereby possibly making another client's or the server's rpm database inconsistent with what's on-disk. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
http://bugzilla.novell.com/show_bug.cgi?id=594223
http://bugzilla.novell.com/show_bug.cgi?id=594223#c6
--- Comment #6 from Dr. Werner Fink
How well does that even integrate with rpm? Updating a client would mean either attempting to write to a read-only /usr, or overwriting the files in /usr, thereby possibly making another client's or the server's rpm database inconsistent with what's on-disk.
You can not maintaine such a cluster on a system based scheme, for such setups you need to share any other partitions. After an update the local configuration for the clients have to be distributed. Maybe with other NFS distributed partitions for every client or with the help of rsync and a script updating the clients local configuration. For the first case on the servers side the mount option --bind and --rbind are very useful. Btw: you way $network -> $networkplus introduce a loop as sshd requires not only /usr but also plain network. For a fix the line # X-Start-Before: +clamd +irq_balancer should be enough. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
http://bugzilla.novell.com/show_bug.cgi?id=594223
http://bugzilla.novell.com/show_bug.cgi?id=594223#c7
--- Comment #7 from Jan Engelhardt
Btw: you way $network -> $networkplus introduce a loop
It does not. Perhaps it was a little unclear; I meant changing $network to $networkplus in /etc/init.d/{almost all services, like clamd}, not insserv.conf. digraph { clam -> networkplus; networkplus -> sshd; sshd -> remote_fs; remote_fs -> nfs; nfs -> network; }; -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
http://bugzilla.novell.com/show_bug.cgi?id=594223
http://bugzilla.novell.com/show_bug.cgi?id=594223#c
Anna Bernathova
http://bugzilla.novell.com/show_bug.cgi?id=594223
http://bugzilla.novell.com/show_bug.cgi?id=594223#c8
Dr. Werner Fink
http://bugzilla.novell.com/show_bug.cgi?id=594223
http://bugzilla.novell.com/show_bug.cgi?id=594223#c9
--- Comment #9 from Jan Engelhardt
https://bugzilla.novell.com/show_bug.cgi?id=594223
https://bugzilla.novell.com/show_bug.cgi?id=594223#c10
--- Comment #10 from Jan Engelhardt
https://bugzilla.novell.com/show_bug.cgi?id=594223
https://bugzilla.novell.com/show_bug.cgi?id=594223#c11
--- Comment #11 from Dr. Werner Fink
https://bugzilla.novell.com/show_bug.cgi?id=594223
https://bugzilla.novell.com/show_bug.cgi?id=594223#c12
--- Comment #12 from Jan Engelhardt
https://bugzilla.novell.com/show_bug.cgi?id=594223
https://bugzilla.novell.com/show_bug.cgi?id=594223#c
Jan Engelhardt
https://bugzilla.novell.com/show_bug.cgi?id=594223
https://bugzilla.novell.com/show_bug.cgi?id=594223#c13
Jan Engelhardt
participants (1)
-
bugzilla_noreply@novell.com