Login or Register to Ask a Question and Join Our Community


Solaris

T2000 - devfsadm problem

Login to Discuss or Reply to this Discussion in Our Community

 
Thread Tools Search this Thread
Operating Systems Solaris T2000 - devfsadm problem
# 8  
Old 01-21-2009
incredible --> I have tried devfsadm -C -c disk and nothing changed
The output of iostat -En |grep Hard is:

Quote:
> iostat -En |grep Hard
c5t5d0 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c0t0d0 Soft Errors: 2 Hard Errors: 0 Transport Errors: 0
c1t0d0 Soft Errors: 4 Hard Errors: 0 Transport Errors: 4
c5t5d1 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c5t6d0 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c5t6d1 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c5t7d0 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c5t7d1 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t7d2 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t4d2 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c5t4d2 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t5d2 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c5t5d2 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t6d2 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c5t6d2 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c5t7d2 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t5d3 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t4d0 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t4d1 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t5d0 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t5d1 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t6d0 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t6d1 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t7d0 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t7d1 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c5t4d0 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c5t4d1 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c5t7d3 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t4d3 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c5t4d3 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c5t6d3 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t6d3 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t7d3 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c5t5d3 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t4d4 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c5t4d4 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c5t5d4 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c5t6d4 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t5d4 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t6d4 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t7d4 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c5t7d4 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t4d5 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t5d5 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t6d5 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c4t7d5 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c5t4d5 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c5t5d5 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c5t6d5 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
c5t7d5 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
And ls /dev/dsk does not have any ddevice of LUN 6 (Which is the LUN that has been assigned to the new disk from the SAN)
The problem with a reconfiguration reboot is that this server is in a production environment and can not be rebooted easily.

Diabolist--> Yes, all other T2000's are connected to the SAN and working without any problems and have been setup with the exact same drivers for the emulex cards. Some of them connect to other SAN's also and have no problems..

Quote:
> cat /etc/path_to_inst |grep emlx
"/pci@7c0/pci@0/pci@8/fibre-channel@0" 0 "emlxs"
"/pci@7c0/pci@0/pci@9/fibre-channel@0" 1 "emlxs"
The output of modinfo:
Quote:
> modinfo |grep eml (Nothing was found)
> modinfo |grep lpf
55 7be00000 82fd8 154 1 lpfc (Emulex FC SCSI/IP lpfc 6.20j)
(the lpfc driver shown here is probably because we install the emulex driver)

During a reconfig reboot (can't do one now), when I have added a new disk, it locates the disk and I think shows messages like "corrupted label on cXtXdX.." for each path of the disk found, (which is correct since it is an new disk and has not been labeled yet).

And yes, we have installed the emulex driver, not only on this server, but on any server where we use emulex cards (about 50 servers) and have never since now had a problem. Altough we have had problems in the past with the Sun emulex driver. (Thats why we decided to use the emulex driver)

Anything else you need, please ask..
Thank you very much for your replies!
# 9  
Old 01-21-2009
Mmm, I will probably be of limited help then.

If you have the same config running on another host, the best suggestion I can give is compare the two. Which I'm sure you've already done now until your eyes bleed Smilie

If you're using the emulex driver, then you can't use cfgadm. The controllers wont show up.

What storage are you using? Compare /kernel/drv/lpfc.conf, sd.conf (if persistent binding). You seem to have some connectivity to the storage, but re-check your lun mapping/masking.

Did you install the emulex toolkit (HBAnywhere?).

Getting devices to dynamically pop up in Solaris doesn't always work too well when not using the leadville stack. It's always a crap shoot on our older systems using qlogic branded qlogic cards (yay for Solaris 10! Sorry, big fan of the Sun driver Smilie )

Have support? Give them a ring... but let us know if you fix it Smilie
# 10  
Old 01-30-2009
I have compared the config files with other working systems, but nothing seamed wrong.

OK, thanks, I didn't know that about cfgadm and Emulex Drivers.

The storage that is connected is an IBM DS4800 connected through IBM fiber optic switches.

I have copied sd.conf and lpfc.conf from working systems to this system and have checked the lun mapping.

Also, I have installed HBAnywhere (all emulex tools, and also have done a firmware upgrade on the emulex cards)

We don't have support for these kind of issues, so any other help would be appreciated.

Thanks again..
# 11  
Old 01-30-2009
check the lpfc.conf file for thw WWPN and see if the correct controller si pointing to the right WWN
# 12  
Old 01-30-2009
lpfc.conf was OK.

It seems to be something with the card or the fiber switch.
I tried resetting the emulex card, using lputil.
When I had reseted both lpfc devices, all the attached disks where found. I did a reset of both emulex cards and only then did the system detect all the paths.

It's not perfect, but for now it works. Smilie
The same thing happend when disabling/enabling the fiber switch port.

I would like to solve this problem though, so if anyone get's any ideas as to what is going on, they will be appreciated..Smilie


(I am going to try and change the emulex cards once I get a pair of new ones..
Thats the only thing I have thought of trying at this point.)
# 13  
Old 01-30-2009
great! good news though but not a right method if this would suddenly occur into a productional situation. will let you know if I can find any clues
# 14  
Old 01-30-2009
Computer
SmilieSmilie
Login or Register to Ask a Question

Previous Thread | Next Thread

10 More Discussions You Might Find Interesting

1. Solaris

Sun Fire 480 - devfsadm -C not working after replacing one of the boot disks

Our SunFire 480 - one of the boot disks failed. It is mirrored with VxVm. I followed the exact steps that is detailed in the veritas and sun guides (I can not post the URL since I don't have 5 posts yet.) After removing the failed disk, I ran devfsadm -C -c disk and it still shows the device... (7 Replies)
Discussion started by: jtamminen
7 Replies

2. Solaris

dvd problem in T2000

hi all when i trying to give auto-boot value as false as following #eeprom auto-boot?=false it is showing as zsh: no matches found: auto-boot?=false so felt this is because of error in hard ware so i tried this #iostat -en it is showing like iostat -en ---- errors --- s/w h/w trn... (4 Replies)
Discussion started by: wkbn86
4 Replies

3. Solaris

Raid problem on T2000

Hi all, First post so be gentile :) It's my first time installing Solaris on T2000 server. I have 2 72GB HD's on it. After configuring HW raid and running raidctl command I get this output: bash-3.00# raidctl -l c0t0d0 Volume Size Stripe Status Cache RAID ... (2 Replies)
Discussion started by: Obi-Wan
2 Replies

4. Solaris

T2000 Server cannot power on

Hi All, I recently had issues with my new T2000 server. I purchased a new mainboard and swapped the CPU and DIMMs to the new board. Now when i power on the server im unable to do so. I get the following error: ‘SC Alert: Host system poweron failed due to fault at MB/FF_POK.' When i do... (2 Replies)
Discussion started by: Caully
2 Replies

5. Solaris

Auto boot problem on Sun T2000

Hi All I have a problem on T2000 server. bash-3.00# uname -a SunOS 5.10 aaa Generic_120011-14 sun4v sparc SUNW,Netra-T2000After the initial solaris installation server failed to boot: Rebooting with command: boot Boot device: disk:b File and args: The file just loaded does not appear... (1 Reply)
Discussion started by: ouzist
1 Replies

6. Solaris

SF T2000 prtdiag problem

Hi, a couple of T2000 servers did not show the component status and the output was too short. There was a patch that needed to be installed. I also updated to the latest recommended patch cluster and OBP patch. After reboot/restart of picl, it worked ok for a couple of weeks. Then suddenly it went... (12 Replies)
Discussion started by: incredible
12 Replies

7. Solaris

Netra T2000 /var/adm/messages problem

Hi, I have a netra t2000 running solaris 10. It is fully patched with the latest patch cluster and also patch 118833-36 and 126897-02. The SC sys_event level is set to 3. I am running a test to show hardware log status, e.g simulating a power supply failure.Hence I unplug one of te... (3 Replies)
Discussion started by: ulemsee
3 Replies

8. Linux

devfsadm equivalent(sol) command in linux os

Hi frnds, what is the devfsadm(solaris) equivalent command in linux os? Thanks in Adv. Arjun Reddy. (2 Replies)
Discussion started by: arjunreddy3
2 Replies

9. Solaris

t2000 DNS issue

I have an issue with DNS on a solaris t2000 When I set up the zones there was an issue with the /etc/resolv.conf, I had made a typo in my DNS server. I only noticed this after creating my new zones. I have corrected this in the global zone and I am now able to ping and nslookup inthe... (2 Replies)
Discussion started by: james dewitt
2 Replies

10. Solaris

T2000 performance issue

Hi, I have a T2000 box connected to very slow disks (EMC SAN ATA disks) by a double port hba through a fabric topology. RAID 5 (6+1) configuration. The issue is using RMAN for backup and validation psent almost twice the time than with a V880 Fire connected the same way. As with the... (2 Replies)
Discussion started by: freezer91
2 Replies
Login or Register to Ask a Question