Hardy beta boot sequence: Network mounts in fstab is done before network is ready.

[Dan Näsman]

Network mounts in fstab is done before network is ready.

I have redirected documents, desktop etc to samba share on a NAS. After logging in it takes about 30 seconds to open documents on the mounted samba share. My guess is that the mount failed during boot and therefore it takes longer when I am opening the mount.

Looking at the kernel loggs for the boot sequence is seems like the cifs mount in fstab is done before the ethernet port is ready. Se the following:
...
Mar 29 14:04:57 Ubuntu-1 kernel: [ 27.339104] lp0: using parport0 (interrupt-driven).

Mar 29 14:04:57 Ubuntu-1 kernel: [ 27.434308] Adding 1646620k swap on /dev/sda5. Priority:-1 extents:1 across:1646620k

Mar 29 14:04:57 Ubuntu-1 kernel: [ 28.073536] EXT3 FS on sda1, internal journal

Mar 29 14:04:57 Ubuntu-1 kernel: [ 29.363556] ip_tables: (C) 2000-2006 Netfilter Core Team

Mar 29 14:04:57 Ubuntu-1 kernel: [ 29.721116] CIFS VFS: Error connecting to IPv4 socket. Aborting operation

Mar 29 14:04:57 Ubuntu-1 kernel: [ 29.721126] CIFS VFS: cifs_mount failed w/return code = -101

Mar 29 14:04:57 Ubuntu-1 kernel: [ 29.734240] CIFS VFS: Error connecting to IPv4 socket. Aborting operation

Mar 29 14:04:57 Ubuntu-1 kernel: [ 29.734246] CIFS VFS: cifs_mount failed w/return code = -101

Mar 29 14:04:57 Ubuntu-1 kernel: [ 29.735536] CIFS VFS: Error connecting to IPv4 socket. Aborting operation

Mar 29 14:04:57 Ubuntu-1 kernel: [ 29.735541] CIFS VFS: cifs_mount failed w/return code = -101

Mar 29 14:04:57 Ubuntu-1 kernel: [ 29.736675] CIFS VFS: Error connecting to IPv4 socket. Aborting operation

Mar 29 14:04:57 Ubuntu-1 kernel: [ 29.736680] CIFS VFS: cifs_mount failed w/return code = -101

Mar 29 14:04:57 Ubuntu-1 kernel: [ 29.744555] CIFS VFS: Error connecting to IPv4 socket. Aborting operation

Mar 29 14:04:57 Ubuntu-1 kernel: [ 29.744563] CIFS VFS: cifs_mount failed w/return code = -101

Mar 29 14:04:57 Ubuntu-1 kernel: [ 29.746131] CIFS VFS: Error connecting to IPv4 socket. Aborting operation

Mar 29 14:04:57 Ubuntu-1 kernel: [ 29.746136] CIFS VFS: cifs_mount failed w/return code = -101

Mar 29 14:04:57 Ubuntu-1 kernel: [ 29.990355] No dock devices found.

Mar 29 14:04:57 Ubuntu-1 kernel: [ 30.062690] ACPI: \_SB_.PCI0.IDE1.PRI1.MAS1: found ejectable bay

Mar 29 14:04:57 Ubuntu-1 kernel: [ 30.062698] ACPI: \_SB_.PCI0.IDE1.PRI1.MAS1: Adding notify handler

Mar 29 14:04:57 Ubuntu-1 kernel: [ 30.062750] ACPI: Bay [\_SB_.PCI0.IDE1.PRI1.MAS1] Added

Mar 29 14:04:58 Ubuntu-1 kernel: [ 31.586003] apm: BIOS version 1.2 Flags 0x03 (Driver version 1.16ac)

Mar 29 14:04:58 Ubuntu-1 kernel: [ 31.586010] apm: disabled - APM is not SMP safe.

Mar 29 14:04:58 Ubuntu-1 kernel: [ 31.726293] ppdev: user-space parallel port driver

Mar 29 14:05:00 Ubuntu-1 kernel: [ 33.433817] Bluetooth: Core ver 2.11

Mar 29 14:05:00 Ubuntu-1 kernel: [ 33.434585] NET: Registered protocol family 31

Mar 29 14:05:00 Ubuntu-1 kernel: [ 33.434590] Bluetooth: HCI device and connection manager initialized

Mar 29 14:05:00 Ubuntu-1 kernel: [ 33.434595] Bluetooth: HCI socket layer initialized

Mar 29 14:05:00 Ubuntu-1 kernel: [ 33.544174] Bluetooth: L2CAP ver 2.9

Mar 29 14:05:00 Ubuntu-1 kernel: [ 33.544180] Bluetooth: L2CAP socket layer initialized

Mar 29 14:05:00 Ubuntu-1 kernel: [ 33.654475] Bluetooth: RFCOMM socket layer initialized

Mar 29 14:05:00 Ubuntu-1 kernel: [ 33.654491] Bluetooth: RFCOMM TTY layer initialized

Mar 29 14:05:00 Ubuntu-1 kernel: [ 33.654494] Bluetooth: RFCOMM ver 1.8

Mar 29 14:05:03 Ubuntu-1 kernel: [ 36.314510] tg3: eth0: Link is up at 1000 Mbps, full duplex.

Mar 29 14:05:03 Ubuntu-1 kernel: [ 36.314515] tg3: eth0: Flow control is on for TX and on for RX.

...

This also seems to go for wifi networks where it takes even longer for the network to be ready.

Regards, Dan

[stu (sgeorge)]

I'm seeing this as well. Also 9 times out of 10 the system hangs on shutdown with hal erros with the CIFS / VFS thing. last apt update was april 12th and its still present for me.

[Dan Näsman (dan-naesman)]

I also see the shutdown hang with "CIFS...". It appeared already in Gutsy when I changed the mounts in fstab from smbfs to cifs.

[Peter Schachte (schachte-csse)]

I see this too on my slow (500 MHz P3, 256 MB memory) Hardy box (Mythbuntu). It hangs for about 30 - 45 seconds on every boot with a message that the network is inaccessible. This problem appeared when I added some CIFS mounts to my /etc/fstab. After the delay, the continues with the boot, and after boot-up, the mounts have in fact happened.

I also was seeing the even longer hang at shutdown mentioned in a comment to this bug report, but found a solution on the Ubuntu forums. Just do:

      sudo ln -s /etc/init.d/umountnfs.sh /etc/rc0.d/K15umountnfs.sh
      sudo ln -s /etc/init.d/umountnfs.sh /etc/rc6.d/K15umountnfs.sh

and that fixes it completely (but doesn't help the delay at start-up).

[Dan Näsman (dan-naesman)]

Thanks a lot. It fixed the shut-down-hang on both my Hardy-PC and Fiesty-PC.

[Litrik De Roy (litrik)]

Is any of you using the "_netdev" mount parameter in /etc/fstab?
This thread indicates that should avoid the delay during boot: http://ubuntuforums.org/showthread.php?t=788146

[Dan Näsman (dan-naesman)]

No.

I use a "cifs" network share. As I understand "_netdev" is only valid
for nfs.

Regards, Dan

Litrik De Roy skrev:
> Is any of you using the "_netdev" mount parameter in /etc/fstab?
> This thread indicates that should avoid the delay during boot: http://ubuntuforums.org/showthread.php?t=788146
>
>

[jtniehof (jtniehof)]

Per the mount manpage, _netdev is a filesystem-independent mount option. Is anyone using _netdev and still having any problems?

[Launchpad Janitor (janitor)]

[Expired for sysvinit (Ubuntu) because there has been no activity for 60 days.]