Mint 14 64bit fresh install on software raid1 loops forever

[alexaandru]

Hello,

I just installed Mind 14 fresh (MATE, 64bit edition), on an existing software RAID1 partition. The install process went like this:

1. boot live cd;
2. install mdadm; run mdadm --assemble --scan;
3. start install process; select existing md partitions (one for root, v0.9x so that grub can see it/install on it and a separate one for home); format the root one and let it install everything fresh;
4. at "setting up documents" or something it throws an error (on the logs I saw python crashed due to not finding ~/.face folder) but says it can continue and it does;
5. at the end I select "continue testing" over reboot and perform the following steps:
6. sudo -i; cd /; for i in proc sys dev dev/pts; do mount --bind $i /target/$i; done; chroot /target; apt-get install mdadm; apt-get install --reinstall grub-pc
7. reboot

And now the problem:

The system boots but enters what it seems an infinite loop and never gets out of it. Hitting ESC, shows the following:

fsck from util-linux 2.20.1
fsck from util-linux 2.20.1
/dev/md1: clean, 175530/525200 files, 938186/2099184 blocks
/dev/md3: clean, 308716/3139584 files, 4920305/12553966 blocks
* Starting bluetooth daemon [ OK ]
* Starting CUPS printing spooler/server [ OK ]
* Starting Samba Auto-reload Integration [ OK ]

it waits for ~30 seconds, then displays the same thing again. And it keeps of cycling this over and over. I stayed once (tried it multiple times, same thing) for 3 minutes and it displayed the same thing like 5 or 6 times. It never gets past it. It does respond to Ctrl+Alt+Del and reboots nicely though.

Here is an excerpt from syslog

Nov 23 22:00:10 t420 kernel: [ 3.457927] sd 6:0:0:0: >Attached scsi generic sg2 type 0
Nov 23 22:00:10 t420 kernel: [ 3.458994] sd 6:0:0:0: >[sdc] 976773168 512-byte logical blocks: (500 GB/465 GiB)
Nov 23 22:00:10 t420 kernel: [ 3.459502] sd 6:0:0:0: >[sdc] Write Protect is off
Nov 23 22:00:10 t420 kernel: [ 3.459504] sd 6:0:0:0: >[sdc] Mode Sense: 03 00 00 00
Nov 23 22:00:10 t420 kernel: [ 3.459997] sd 6:0:0:0: >[sdc] No Caching mode page present
Nov 23 22:00:10 t420 kernel: [ 3.459998] sd 6:0:0:0: >[sdc] Assuming drive cache: write through
Nov 23 22:00:10 t420 kernel: [ 3.462547] sd 6:0:0:0: >[sdc] No Caching mode page present
Nov 23 22:00:10 t420 kernel: [ 3.462549] sd 6:0:0:0: >[sdc] Assuming drive cache: write through
Nov 23 22:00:10 t420 kernel: [ 3.463180] sdc: sdc1
Nov 23 22:00:10 t420 kernel: [ 3.476330] sd 6:0:0:0: >[sdc] No Caching mode page present
Nov 23 22:00:10 t420 kernel: [ 3.476333] sd 6:0:0:0: >[sdc] Assuming drive cache: write through
Nov 23 22:00:10 t420 kernel: [ 3.476335] sd 6:0:0:0: >[sdc] Attached SCSI disk
Nov 23 22:00:10 t420 kernel: [ 3.539056] EXT4-fs (md1): mounted filesystem with ordered data mode. Opts: (null)
Nov 23 22:00:10 t420 kernel: [ 4.132596] Adding 4808700k swap on /dev/sda6. Priority:-1 extents:1 across:4808700k
Nov 23 22:00:10 t420 kernel: [ 4.901871] EXT4-fs (md1): re-mounted. Opts: errors=remount-ro
Nov 23 22:00:10 t420 kernel: [ 4.956237] EXT4-fs (md3): mounted filesystem with ordered data mode. Opts: (null)
Nov 23 22:00:10 t420 kernel: [ 5.927159] IPv6: ADDRCONF(NETDEV_UP): eth0: link is not ready
Nov 23 22:00:10 t420 kernel: [ 6.132027] Bluetooth: Core ver 2.16
Nov 23 22:00:10 t420 kernel: [ 6.132042] NET: Registered protocol family 31
Nov 23 22:00:10 t420 kernel: [ 6.132044] Bluetooth: HCI device and connection manager initialized
Nov 23 22:00:10 t420 kernel: [ 6.132046] Bluetooth: HCI socket layer initialized
Nov 23 22:00:10 t420 kernel: [ 6.132047] Bluetooth: L2CAP socket layer initialized
Nov 23 22:00:10 t420 kernel: [ 6.132051] Bluetooth: SCO socket layer initialized
Nov 23 22:00:10 t420 kernel: [ 6.136818] Bluetooth: RFCOMM TTY layer initialized
Nov 23 22:00:10 t420 kernel: [ 6.136822] Bluetooth: RFCOMM socket layer initialized
Nov 23 22:00:10 t420 kernel: [ 6.136823] Bluetooth: RFCOMM ver 1.11
Nov 23 22:00:10 t420 kernel: [ 6.145806] lp: driver loaded but no devices found
Nov 23 22:00:10 t420 kernel: [ 6.147633] Bluetooth: BNEP (Ethernet Emulation) ver 1.3
Nov 23 22:00:10 t420 kernel: [ 6.147634] Bluetooth: BNEP filters: protocol multicast
Nov 23 22:00:10 t420 kernel: [ 6.167770] ppdev: user-space parallel port driver
Nov 23 22:02:26 t420 kernel: Kernel logging (proc) stopped.
Nov 23 22:02:26 t420 rsyslogd: [origin software="rsyslogd" swVersion="5.8.6" x-pid="502" x-info="http://www.rsyslog.com"] exiting on signal 15.

As you can see it matches what's in boot - load bluetooth, load ppdev which is related to CUPS and then load nothing else for 2 minutes (during which time it displayed the same bluetooth + cups + samba messages) until it shuts off due to Ctrl+Alt+Del.

Booting the live CD works like a charm (I'm using it to write this post...). This is not my first Mint install, I'm a happy Mint user since version 11 I think (at the time when Ubuntu made its distro truly umbearable). I use software raid ever since I can remember (been a Linux user for 12 years now) and anyway, booting from the raid partition does not seem to be a problem. I just have no clue what to try next (other than reverting to Mint 13 which worked like a charm). Any hints are greatly appreciated.

Cheers!
Alex

[alexaandru]

Seeing that the last service it was trying to load was Samba, I removed it and gave another boot a shot. Here's the boot.log now:

fsck from util-linux 2.20.1
fsck from util-linux 2.20.1
/dev/md1: clean, 175348/525200 files, 919146/2099184 blocks
/dev/md3: clean, 308716/3139584 files, 4920305/12553966 blocks
process 393: arguments to dbus_server_disconnect() were incorrect, assertion "old_refcount > 0" failed in file ../../dbus/dbus-server.c line 786.
This is normally a bug in some application using the D-Bus library.
process 393: arguments to dbus_server_unref() were incorrect, assertion "old_refcount > 0" failed in file ../../dbus/dbus-server.c line 749.
This is normally a bug in some application using the D-Bus library.
* Starting mDNS/DNS-SD daemon [ OK ]
* Starting bluetooth daemon [ OK ]
* Starting CUPS printing spooler/server [ OK ]

and then the same loop. So it wasn't related to Samba at least.

Here's the syslog excerpt:

Nov 23 22:35:53 t420 kernel: [ 6.187035] lp: driver loaded but no devices found
Nov 23 22:35:53 t420 kernel: [ 6.189411] ppdev: user-space parallel port driver
Nov 23 22:35:53 t420 avahi-daemon[551]: Found user 'avahi' (UID 111) and group 'avahi' (GID 120).
Nov 23 22:35:53 t420 avahi-daemon[551]: Successfully dropped root privileges.
Nov 23 22:35:53 t420 avahi-daemon[551]: avahi-daemon 0.6.31 starting up.
Nov 23 22:35:53 t420 avahi-daemon[551]: Successfully called chroot().
Nov 23 22:35:53 t420 avahi-daemon[551]: Successfully dropped remaining capabilities.
Nov 23 22:35:53 t420 avahi-daemon[551]: Loading service file /services/udisks.service.
Nov 23 22:35:53 t420 avahi-daemon[551]: Network interface enumeration completed.
Nov 23 22:35:53 t420 avahi-daemon[551]: Registering HINFO record with values 'X86_64'/'LINUX'.
Nov 23 22:35:53 t420 avahi-daemon[551]: Server startup complete. Host name is t420.local. Local service cookie is 612887680.
Nov 23 22:35:53 t420 avahi-daemon[551]: Service "t420" (/services/udisks.service) successfully established.
Nov 23 22:35:53 t420 bluetoothd[553]: Bluetooth daemon 4.101
Nov 23 22:35:53 t420 kernel: [ 6.283918] Bluetooth: Core ver 2.16
Nov 23 22:35:53 t420 kernel: [ 6.283935] NET: Registered protocol family 31
Nov 23 22:35:53 t420 kernel: [ 6.283937] Bluetooth: HCI device and connection manager initialized
Nov 23 22:35:53 t420 kernel: [ 6.283938] Bluetooth: HCI socket layer initialized
Nov 23 22:35:53 t420 kernel: [ 6.283939] Bluetooth: L2CAP socket layer initialized
Nov 23 22:35:53 t420 kernel: [ 6.283942] Bluetooth: SCO socket layer initialized
Nov 23 22:35:53 t420 kernel: [ 6.286383] Bluetooth: RFCOMM TTY layer initialized
Nov 23 22:35:53 t420 kernel: [ 6.286387] Bluetooth: RFCOMM socket layer initialized
Nov 23 22:35:53 t420 kernel: [ 6.286388] Bluetooth: RFCOMM ver 1.11
Nov 23 22:35:53 t420 bluetoothd[553]: Starting SDP server
Nov 23 22:35:53 t420 bluetoothd[553]: DIS cannot start: GATT is disabled
Nov 23 22:35:53 t420 bluetoothd[553]: Failed to init deviceinfo plugin
Nov 23 22:35:53 t420 bluetoothd[553]: Failed to init proximity plugin
Nov 23 22:35:53 t420 bluetoothd[553]: Failed to init time plugin
Nov 23 22:35:53 t420 bluetoothd[553]: Failed to init alert plugin
Nov 23 22:35:53 t420 bluetoothd[553]: Failed to init thermometer plugin
Nov 23 22:35:53 t420 kernel: [ 6.303358] Bluetooth: BNEP (Ethernet Emulation) ver 1.3
Nov 23 22:35:53 t420 kernel: [ 6.303362] Bluetooth: BNEP filters: protocol multicast
Nov 23 22:35:53 t420 bluetoothd[553]: Failed to init gatt_example plugin
Nov 23 22:35:53 t420 bluetoothd[553]: Bluetooth Management interface initialized

Gonna remove bluetooth next... Would be nice if there was a way to interactively select services on boot (the way RedHat used to allow - not sure if it still does).

[alexaandru]

I am back to MATE 13 64bit (same story: software raid setup). Total install time including the manual raid setup steps? 8 minutes. And it all works just fine.

[tcope]

I too am experiencing similar issues with Mate14 but on a 32 bit install. Previous versions of mint
seem to have better raid support. There seems to be some wonky thing happening with the relationship
between mdadm and device mapper. Finally after using a blow torch and a jack hammer raid is setup
and seems to be booting then blam .. same type error as this in this thread. Only difference is that its
not looping, just stops.
/boot = raid1
/ = raid10

It's not like raid is some new magical thing. It's well established, and in hi demand.
Proper support for it should be standard. I have read posts around that the server
kernels may have better raid support. I don't know where the mind set got fixed
that only servers need raid. I think it would be a safe bet to say that many if not
most users of hi performance work stations utilize some form of raid.

Would rather not walk backwards. Will keep looking for a solution.

[mintybits]

While in chroot you may want to make sure fstab is correct and make an entry in /etc/mdadm/mdadm.conf for your array and then run

Code: Select all

sudo update-initramfs -u

[mintybits]

3. start install process; select existing md partitions (one for root, v0.9x so that grub can see it/install on it and a separate one for home); format the root one and let it install everything fresh;

Grub can boot the latest mdadm format. I recommend using 1.2 although I don't think this affects the bootability in this case.

[tcope]

The error that is occuring is after grub2. Had been previously running "update-initramfs -u" .
Still have not found the cause of the issue. Have even tried a slightly newer kernel without luck.
Still looking for a solution.

[tcope]

After wasting over 40 hours on this issue, I have gotten no where.
Nothing I try has made any difference nor has produces any relevant
error message. The boot just stops and thats it. The system is alive because
plugging a usb stick produces the apropriate system message on the screen
but you have no way to interact with it. Every thing looks fine, it just wont finish
booting. I too am regressing to Mint 13 till this issue can be fixed.

[blackmole]

Same problem here on Mint 14 Cinnamon 64bit. But after a few hours, I was able to solve it.

First of all, the loop mentioned is no loop, it's just that the output is added everytime when switching from the splash image to the console output, a "clear" is missing there.

However, the error results from the fake start-stop-daemon and the fake initctl. To solve it, do a chroot into the new system (e.g., from the live cd) and simplely do a

Code: Select all

apt-get install --reinstall dpkg upstart

I'm not sure why these fake programs are there, but it is pretty annoying.