[Bug 1146698] New: wicked breaks NFS root
http://bugzilla.suse.com/show_bug.cgi?id=1146698 Bug ID: 1146698 Summary: wicked breaks NFS root Classification: openSUSE Product: openSUSE Tumbleweed Version: Current Hardware: Other OS: Other Status: NEW Severity: Normal Priority: P5 - None Component: Network Assignee: wicked-maintainers@suse.de Reporter: schwab@suse.com QA Contact: qa-bugs@suse.de Found By: --- Blocker: --- When running with NFS root the network interface must never be brought down, not even momentarily. wicked is violating that. Starting wicked managed network interfaces... [ *** ] A start job is running for wicked m…etwork interfaces (18s / no limit) [ 223.408007] nfs: server 10.160.4.0 not responding, still trying [ 226.528002] nfs: server 10.160.4.0 not responding, still trying [ 243.977780] nfs: server 10.160.4.0 not responding, still trying [ 243.982918] nfs: server 10.160.4.0 not responding, still trying [ 245.247984] nfs: server 10.160.4.0 not responding, still trying [ 248.368005] nfs: server 10.160.4.0 not responding, still trying [ 251.488000] nfs: server 10.160.4.0 not responding, still trying [ 254.608004] nfs: server 10.160.4.0 not responding, still trying [ 257.727984] nfs: server 10.160.4.0 not responding, still trying [ 260.847989] nfs: server 10.160.4.0 not responding, still trying [ 263.967983] nfs: server 10.160.4.0 not responding, still trying [ 267.087987] nfs: server 10.160.4.0 not responding, still trying [ 269.327783] nfs: server 10.160.4.0 not responding, timed out [ 270.207990] nfs: server 10.160.4.0 not responding, still trying [ 273.327979] nfs: server 10.160.4.0 not responding, still trying [ 276.447989] nfs: server 10.160.4.0 not responding, still trying [ 279.568007] nfs: server 10.160.4.0 not responding, still trying [ 282.687976] nfs: server 10.160.4.0 not responding, still trying -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.suse.com/show_bug.cgi?id=1146698
http://bugzilla.suse.com/show_bug.cgi?id=1146698#c1
Marius Tomaschewski
http://bugzilla.suse.com/show_bug.cgi?id=1146698
http://bugzilla.suse.com/show_bug.cgi?id=1146698#c2
Andreas Schwab
http://bugzilla.suse.com/show_bug.cgi?id=1146698
http://bugzilla.suse.com/show_bug.cgi?id=1146698#c3
Rubén Torrero Marijnissen
http://bugzilla.suse.com/show_bug.cgi?id=1146698
http://bugzilla.suse.com/show_bug.cgi?id=1146698#c4
Marius Tomaschewski
http://bugzilla.suse.com/show_bug.cgi?id=1146698
http://bugzilla.suse.com/show_bug.cgi?id=1146698#c5
--- Comment #5 from Marius Tomaschewski
http://bugzilla.suse.com/show_bug.cgi?id=1146698
Marius Tomaschewski
http://bugzilla.suse.com/show_bug.cgi?id=1146698
http://bugzilla.suse.com/show_bug.cgi?id=1146698#c6
Andreas Schwab
http://bugzilla.suse.com/show_bug.cgi?id=1146698
http://bugzilla.suse.com/show_bug.cgi?id=1146698#c7
--- Comment #7 from Andreas Schwab
http://bugzilla.suse.com/show_bug.cgi?id=1146698
http://bugzilla.suse.com/show_bug.cgi?id=1146698#c8
Marius Tomaschewski
eth0 up link: #2, state up, mtu 1500 type: ethernet, hwaddr 70:b3:d5:92:f1:07 control: persistent ^^^^^^^^^^^^^^^^^^^^
OK, it is marked nfsroot. When you call "ifdown eth0" or "rcnetwork stop" now, you'll see that wicked does not perform any ifdown happens, but this message is printed: wicked: skipping eth0 interface: persistent mode is on For anything else, please attach debug logs as described in comment 3. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.suse.com/show_bug.cgi?id=1146698
http://bugzilla.suse.com/show_bug.cgi?id=1146698#c9
Andreas Schwab
http://bugzilla.suse.com/show_bug.cgi?id=1146698
http://bugzilla.suse.com/show_bug.cgi?id=1146698#c10
Marius Tomaschewski
http://bugzilla.suse.com/show_bug.cgi?id=1146698
http://bugzilla.suse.com/show_bug.cgi?id=1146698#c11
Andreas Schwab
http://bugzilla.suse.com/show_bug.cgi?id=1146698
http://bugzilla.suse.com/show_bug.cgi?id=1146698#c12
Rubén Torrero Marijnissen
http://bugzilla.suse.com/show_bug.cgi?id=1146698
http://bugzilla.suse.com/show_bug.cgi?id=1146698#c13
Rubén Torrero Marijnissen
http://bugzilla.suse.com/show_bug.cgi?id=1146698
http://bugzilla.suse.com/show_bug.cgi?id=1146698#c14
Rubén Torrero Marijnissen
http://bugzilla.suse.com/show_bug.cgi?id=1146698
http://bugzilla.suse.com/show_bug.cgi?id=1146698#c15
--- Comment #15 from Marius Tomaschewski
Then why did it hang?
Probably the servers / network were down. (In reply to Andreas Schwab from comment #0)
When running with NFS root the network interface must never be brought down, not even momentarily. wicked is violating that.
Starting wicked managed network interfaces... [ *** ] A start job is running for wicked m…etwork interfaces (18s / no limit) [ 223.408007] nfs: server 10.160.4.0 not responding, still trying
What you provide here is: wicked is **starting* since 18sec, probably trying to get an IP address and either the NIC does not have carrier or the dhcp + nfs server aren't reachable for another reasons. "network interface must never be brought down" is not the case here as visible in "Starting wicked" and "A start job is running for wicked". Wicked does not execute any "systemctl start / restart network"; this is made either because the service is enabled and requested by some target to start in by the user. (In reply to Andreas Schwab from comment #6)
WTF?
Is also not a helpful information. As no valuable informations about the issue have been provided, just some statement with contradicting "log" output -> INVALID report, or WORKSFORME if you prefer: "The problem described cannot be duplicated. If more information is provided, the bug can be reopened." -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.suse.com/show_bug.cgi?id=1146698
http://bugzilla.suse.com/show_bug.cgi?id=1146698#c16
Andreas Schwab
Probably the servers / network were down.
Nope, otherwise the kernel and initrd could not have been loaded.
"network interface must never be brought down" is not the case here as visible in "Starting wicked" and "A start job is running for wicked".
When the network is brought down, all processes will become stuck in D. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.suse.com/show_bug.cgi?id=1146698
Rubén Torrero Marijnissen
http://bugzilla.suse.com/show_bug.cgi?id=1146698
http://bugzilla.suse.com/show_bug.cgi?id=1146698#c17
Rossella Sblendido
http://bugzilla.suse.com/show_bug.cgi?id=1146698
http://bugzilla.suse.com/show_bug.cgi?id=1146698#c18
--- Comment #18 from Marius Tomaschewski
"network interface must never be brought down" is not the case here as visible in "Starting wicked" and "A start job is running for wicked".
When the network is brought down, all processes will become stuck in D.
But it isn't brought down, but is trying to **start**. Putting this issue to my IGNORE list -- useless ping-pong. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.suse.com/show_bug.cgi?id=1146698
http://bugzilla.suse.com/show_bug.cgi?id=1146698#c19
Andreas Schwab
But it isn't brought down,
Yes, it is. It is _started_ by the kernel. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.suse.com/show_bug.cgi?id=1146698
http://bugzilla.suse.com/show_bug.cgi?id=1146698#c20
Rossella Sblendido
http://bugzilla.suse.com/show_bug.cgi?id=1146698
http://bugzilla.suse.com/show_bug.cgi?id=1146698#c21
Rossella Sblendido
participants (1)
-
bugzilla_noreply@novell.com