| Summary: | NFS mount fails: network.service starts after remote-fs | ||
|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Hans Ecke <hansecke> |
| Component: | systemd | Assignee: | systemd-maint |
| Status: | CLOSED WORKSFORME | QA Contact: | Fedora Extras Quality Assurance <extras-qa> |
| Severity: | unspecified | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | 16 | CC: | fedora, johannbg, metherid, mschmidt, msekleta, notting, plautrba, systemd-maint |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | Bug Fix | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2012-10-30 17:58:17 UTC | Type: | --- |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Attachments: | |||
|
Description
Hans Ecke
2012-01-11 19:56:38 UTC
Created attachment 552214 [details]
output from "systemctl dump"
Created attachment 552215 [details]
output from "/bin/systemd --test --system --log-level=debug 2>&1"
This was done as user "hans". systemd refuses to run this as root.
Created attachment 552216 [details]
a copy of /var/log/messages
... I deleted everything before the bootup
A few comments: 1.) I changed the /etc/init.d/network script with these lines: echo "Running /etc/init.d/network (stderr)" >&2 logger "Running /etc/init.d/network (syslog)" echo "Running /etc/init.d/network (kmsg)" > /dev/kmsg at the very top. You can see the output in dmesg.txt at 95.657956. Obviously the network should have been brought up much earlier.... 2.) systemd unit files are unchanged. /etc/init.d/network is unchanged except for the above 3 lines. 3.) root@bose:/mpl/bose 518> rpm -qa|fgrep NetworkM NetworkManager-gtk-0.9.2-1.fc16.x86_64 NetworkManager-glib-0.9.2-1.fc16.x86_64 4.) root@bose:/mpl/bose 519> systemctl is-enabled network.service network.service is not a native service, redirecting to /sbin/chkconfig. Executing /sbin/chkconfig network --level=5 enabled 5.) root@bose:/mpl/bose 521> systemctl is-enabled remote-fs.target enabled There's a timeout waiting for a non-sensical device unit: [ 93.780946] systemd[1]: Job pavlov:-data.device/start timed out. Do you have a non-zero value in the sixth fstab field (fs_passno) in /etc/fstab for any of the NFS mounts? I think that's what confuses systemd here. That was indeed the case, and now it mounts correctly. Thank you Michal! However, I still get a huge wait/timeout. I'll attach a new dmesg file so you can see it. Any thoughts? Created attachment 552261 [details]
dmesg after setting the fs_passno field of NFS mounts to 0
From the log: [ 24.887032] systemd[1]: About to execute: /etc/rc.d/init.d/network start [ 24.911899] systemd[1]: Forked /etc/rc.d/init.d/network as 971 [ 24.911941] systemd[1]: network.service changed dead -> start [ 24.922464] logger[983]: Running /etc/init.d/network (syslog) [ 24.922569] Running /etc/init.d/network (kmsg) OK, the network script has been spawned. [ 27.470124] systemd[1]: Child 971 belongs to network.service [ 27.470137] systemd[1]: network.service: control process exited, code=exited status=0 The network script finished. [ 27.470142] systemd[1]: network.service got final SIGCHLD for state start [ 27.470193] systemd[1]: network.service changed start -> running [ 27.470206] systemd[1]: Job network.service/start finished, result=done [ 27.491866] systemd[1]: network.target changed dead -> active [ 27.491873] systemd[1]: Job network.target/start finished, result=done Mounting starts at this point. [ 29.360845] systemd[1]: Received SIGCHLD from PID 1439 (ifup-eth). [ 29.360878] systemd[1]: Got SIGCHLD for process 1439 (ifup-eth) [ 29.385851] systemd[1]: Received SIGCHLD from PID 1444 (ifup-eth). [ 29.385882] systemd[1]: Got SIGCHLD for process 1444 (ifup-eth) [ 29.387120] systemd[1]: network.service: cgroup is empty It's odd that the two ifup-eth processes outlived the network script and finished only now. I don't know where this asynchronicity comes from. Are you experiencing the same issue, with latest versions of systemd, initscripts and bash? I consulted this issue with maintainer of initscripts and we didn't figure out how is it possible for /etc/rc.d/init.d/network to exit earlier than ifup-eth script. Hi Michal- I do not see this bug anymore with current package versions. Feel free to close issue and thank you for your work. Hans |