RHEL 4 hang's at boot after migration to EVA 8400

[linslus]

We have problems whit several Redhat linux (RHEL 4.8) when we running to our new EVA 8400 SAN, the servers worked perfect againt our old EVA 4000 SAN.

What happends is that they hang at boot for about 30-60 mintues. It hangs after the "Remounting root filesystem"

Here is a link to the messagelog from the server when it boot's and hangs and then continue:

http://firefly.ibk.se/~linus/linux-boot-san-fail.txt

Here are some rows from the log:

Feb 18 13:55:33 santest01 kernel: Calibrating delay using timer specific routine.. 4409.38 BogoMIPS (lpj=2204692)
Feb 18 13:55:20 santest01 lvm.static: Found duplicate PV Mn9dDJIy3KIxE8upjxZRyt5DdxhaZQ01: using /dev/sdb not /dev/sda


Feb 18 13:55:40 santest01 kernel: qla2300 0000:06:01.1: scsi(1:0:3:1): ABORTing cmd=00000103f8187080 sp=00000103f78a37c0 flags=2 state=7 ext_hist=0 jiffies = 0xfffbf6e0, timeout=7, dpc_flags=4000000, vis_ha->dpc_flags=4000000
q->flag=0 ha=00000103f77103c8 vis_ha=00000103f77103c8 sp->ha=00000103f77103c8
Feb 18 13:55:40 santest01 kernel: qla2xxx_eh_abort(1): aborting sp 00000103f78a37c0 from RISC. pid=1135 sp->state=7 q->q_flag=0

[Amar_Joshi]

Johan,
How about Device mapper? What version are you using?

Right now the option is to remove fiber connection and boot the server without EVA and drop us list of

# lsmod
# modprobe --list
# chkconfig --list
# cat /etc/multipath.conf

[linslus]

It's not possible for me to disconnect fiber at this time, however, there is no problem to boot the server as long there is no disc/LUN presentated to the server but when we present a disc it hangs at boot.

I have attached a textfile whit the output you asked for (at this time one disc is present and the server is up and running)

[Amar_Joshi]

It looks like /etc/multipath.conf is not configured to recognize the EVA devices. Please see attached install/config guide (page8) and edit the /etc/multipath.conf to include:

device {
vendor â HPâ
product "HSV2[01]0|HSV300|HSV4[05]0"
getuid_callout "/sbin/scsi_id -g -u -s /block/%n"
prio_callout "/sbin/mpath_prio_alua /dev/%n"
hardware_handler "0"
path_selector "round-robin 0"
path_grouping_policy group_by_prio
failback immediate
rr_weight uniform
no_path_retry 18
rr_min_io 100
path_checker tur
}

please note that this section goes under devices { ...... }

After that restart the multipath
# service multipathd restart

If everything goes well; you should see the devices and their paths with "multipath -ll" command.

[Amar_Joshi]

PLEASE DON'T copy/paste from my previous post as it has some invalid characters and some lines are combined into single line.

[linslus]

I have added what you said to the multipath.conf. the server still hangs at boot for about 20-45 minutes before it start up. and still same errors in the logs

[root@santest01 ~]# multipath -ll
mpath5 (3600508b4000e300f0000a00003e10000)
[size=1 GB][features="1 queue_if_no_path"][hwhandler="0"]
\_ round-robin 0 [prio=200][active]
\_ 0:0:0:1 sda 8:0 [active][ready]
\_ 0:0:1:1 sdb 8:16 [active][ready]
\_ 1:0:0:1 sde 8:64 [active][ready]
\_ 1:0:1:1 sdf 8:80 [active][ready]
\_ round-robin 0 [prio=40][enabled]
\_ 0:0:3:1 sdd 8:48 [active][ready]
\_ 0:0:2:1 sdc 8:32 [active][ready]
\_ 1:0:2:1 sdg 8:96 [active][ready]
\_ 1:0:3:1 sdh 8:112 [active][ready]

[Johan Guldmyr]

I saw a similar post like this on the forum a while ago.

The problem was that in fstab they had configured an on-boot fsck. Which apparently was not the way to go with SAN-disks.

Do you have that?

[linslus]

The dics is not present in fstab on this server at the moment, so thats is not the case :(

[Amar_Joshi]

There is another section in /etc/multipath.conf to denied devices not part of multipathing configuration.

denied {
devnode "^(ram|raw|loop|fd|md|sr|scd|st)[0-9]*"
devnode "^(dm-0|dm-1)"
devnode "^hd[a-z]"
devnode "^cciss!c[0-9]d[0-9]*"
}

Try adding it into /etc/multipath.conf and see if non-multipath devices are being scanned by mulatipath daemon.

As I am running out of ideas, would like you to check the myvg (root volumegroup):
# vgdisplay -v /dev/myvg

[linslus]

Solved the problem, https://access.redhat.com/kb/docs/DOC-47282