I have an old dell 2450 that was running kernel 2.6.18-308.1.1.el5 I then upgraded it via yum to 2.6.18-308.1.1.el5. Now when the server boots to the new kernel, 2.6.18-308.1.1.el5 it hangs on "starting iSCSI" and the weird thing is when I tried to switch back to the older 2.6.18-308.1.1.el5 kernel it works fine but the network script fails to start. I'm not really sure what I should do, if someone can give me an idea on what I need to do to fix the iSCSId issue on the new kernel or revert back to the old kernel and fix the network issue. The odd thing is both kernel load the e100 network driver but on the older kernel I can get the network script to start.
I would appreciate some help.
paul
On Fri, Apr 13, 2012 at 12:19 PM, Paul A razor@meganet.net wrote:
I have an old dell 2450 that was running kernel 2.6.18-308.1.1.el5 I then upgraded it via yum to 2.6.18-308.1.1.el5. Now when the server boots to the new kernel, 2.6.18-308.1.1.el5 it hangs on "starting iSCSI" and the weird thing is when I tried to switch back to the older 2.6.18-308.1.1.el5 kernel it works fine but the network script fails to start. I'm not really sure what I should do, if someone can give me an idea on what I need to do to fix the iSCSId issue on the new kernel or revert back to the old kernel and fix the network issue. The odd thing is both kernel load the e100 network driver but on the older kernel I can get the network script to start.
I would appreciate some help.
Do you need iscsi? If not, boot with the old kernel, and disable iscsi
# chkconfig --list | grep iscsi (will probably show iscsi and iscsid)
Then # chkconfig iscsi off # chkconfig iscsid off
Then reboot with the new kernel.
I'm running Raid with scsi disks so I'm assuming it's needed correct?
-----Original Message----- From: centos-bounces@centos.org [mailto:centos-bounces@centos.org] On Behalf Of Dale Dellutri Sent: Friday, April 13, 2012 1:32 PM To: CentOS mailing list Subject: Re: [CentOS] upgrade issue
On Fri, Apr 13, 2012 at 12:19 PM, Paul A razor@meganet.net wrote:
I have an old dell 2450 that was running kernel 2.6.18-308.1.1.el5 I then upgraded it via yum to 2.6.18-308.1.1.el5. Now when the server boots to the new kernel, 2.6.18-308.1.1.el5 it hangs on "starting iSCSI" and the weird thing is when I tried to switch back to the older 2.6.18-308.1.1.el5 kernel it works fine but the network script fails to start. I'm not really sure what I should do, if someone can give me an idea on what I need to do to fix the iSCSId issue on the new kernel or revert back to the old kernel and fix the network issue. The odd thing is both kernel load the e100 network driver but on the older kernel
I can get the network script to start.
I would appreciate some help.
Do you need iscsi? If not, boot with the old kernel, and disable iscsi
# chkconfig --list | grep iscsi (will probably show iscsi and iscsid)
Then # chkconfig iscsi off # chkconfig iscsid off
Then reboot with the new kernel.
-- Dale Dellutri _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
On Fri, Apr 13, 2012 at 2:02 PM, Paul A razor@meganet.net wrote:
I'm running Raid with scsi disks so I'm assuming it's needed correct?
iSCSI is for carrying SCSI command s and data over IP networks. I don't know how your RAID is set up, but it isn't normally done with iSCSI. See, for example: http://en.wikipedia.org/wiki/ISCSI versus http://en.wikipedia.org/wiki/RAID
-----Original Message----- From: centos-bounces@centos.org [mailto:centos-bounces@centos.org] On Behalf Of Dale Dellutri Sent: Friday, April 13, 2012 1:32 PM To: CentOS mailing list Subject: Re: [CentOS] upgrade issue
On Fri, Apr 13, 2012 at 12:19 PM, Paul A razor@meganet.net wrote:
I have an old dell 2450 that was running kernel 2.6.18-308.1.1.el5 I then upgraded it via yum to 2.6.18-308.1.1.el5. Now when the server boots to the new kernel, 2.6.18-308.1.1.el5 it hangs on "starting iSCSI" and the weird thing is when I tried to switch back to the older 2.6.18-308.1.1.el5 kernel it works fine but the network script fails to start. I'm not really sure what I should do, if someone can give me an idea on what I need to do to fix the iSCSId issue on the new kernel or revert back to the old kernel and fix the network issue. The odd thing is both kernel load the e100 network driver but on the older kernel
I can get the network script to start.
I would appreciate some help.
Do you need iscsi? If not, boot with the old kernel, and disable iscsi
# chkconfig --list | grep iscsi (will probably show iscsi and iscsid)
Then # chkconfig iscsi off # chkconfig iscsid off
Then reboot with the new kernel.
-- Dale Dellutri _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Dale, I disabled it and the server came up but im seeing all sorts of kernel errors when its disabled ( see below ). If this can't be fixed I would really like to boot the previous kernel however the network script won't start. Im not really sure how upgrading the kernel caused the network to stop working on the previous kernel.
Apr 13 15:21:31 bakup kernel: INFO: task modprobe:740 blocked for more than 120 seconds. Apr 13 15:21:31 bakup kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. Apr 13 15:21:31 bakup kernel: modprobe D 00000010 2276 740 665 (NOTLB) Apr 13 15:21:31 bakup kernel: f78b6b0c 00000086 0f2dbfc1 00000010 f89aa522 f78ea7a4 f78ea2e0 00000006 Apr 13 15:21:31 bakup kernel: f78b5aa0 0f322e2d 00000010 00046e6c 00000001 f78b5bac c2013544 f784d200 Apr 13 15:21:31 bakup kernel: 00000286 f79f98a0 00000000 f7995400 f7dfe4a8 f7dfe4a8 c2013a00 f79f98a0 Apr 13 15:21:31 bakup kernel: Call Trace: Apr 13 15:21:31 bakup kernel: [<f89aa522>] megaraid_queue+0x8ce/0x8d8 [megaraid] Apr 13 15:21:31 bakup kernel: [<c0623b1d>] wait_for_completion+0x6b/0x8f Apr 13 15:21:31 bakup kernel: [<c041f863>] default_wake_function+0x0/0xc Apr 13 15:21:31 bakup kernel: [<c04e52e0>] blk_execute_rq+0x7d/0x97 Apr 13 15:21:31 bakup kernel: [<c04e48a4>] blk_end_sync_rq+0x0/0x1d Apr 13 15:21:31 bakup kernel: [<c045d49c>] mempool_alloc+0x28/0xc9 Apr 13 15:21:31 bakup kernel: [<c04edd12>] cfq_set_request+0x0/0x3cb Apr 13 15:21:31 bakup kernel: [<c04e3132>] blk_recount_segments+0x13/0x21 Apr 13 15:21:31 bakup kernel: [<f8861006>] scsi_execute+0xad/0xbf [scsi_mod] Apr 13 15:21:31 bakup kernel: [<f886109c>] scsi_execute_req+0x84/0xa1 [scsi_mod] Apr 13 15:21:31 bakup kernel: [<f88622ca>] scsi_probe_and_add_lun+0x1da/0x8c1 [scsi_mod] Apr 13 15:21:31 bakup kernel: [<f8862eaa>] __scsi_scan_target+0xb1/0x564 [scsi_mod] Apr 13 15:21:31 bakup kernel: [<c04f1b5b>] kobject_uevent_env+0x3d8/0x3fb Apr 13 15:21:31 bakup kernel: [<c04b20e9>] sysfs_new_dirent+0x53/0x5d Apr 13 15:21:31 bakup kernel: [<c04b213d>] sysfs_make_dirent+0x10/0x66 Apr 13 15:21:31 bakup kernel: [<f886339c>] scsi_scan_channel+0x3f/0x6e [scsi_mod] Apr 13 15:21:31 bakup kernel: [<f886346e>] scsi_scan_host_selected+0xa3/0xd8 [scsi_mod] Apr 13 15:21:31 bakup kernel: [<f88634b2>] scsi_scan_host+0xf/0x12 [scsi_mod] Apr 13 15:21:31 bakup kernel: [<f89ac7ee>] megaraid_probe_one+0xd7e/0xfbc [megaraid] Apr 13 15:21:31 bakup kernel: [<c0563048>] __driver_attach+0x0/0x6b Apr 13 15:21:31 bakup kernel: [<c04fe989>] pci_device_probe+0x36/0x57 Apr 13 15:21:31 bakup kernel: [<c0562f9b>] driver_probe_device+0x54/0xa4 Apr 13 15:21:31 bakup kernel: [<c056308c>] __driver_attach+0x44/0x6b Apr 13 15:21:31 bakup kernel: [<c0562969>] bus_for_each_dev+0x37/0x59 Apr 13 15:21:31 bakup kernel: [<c0562ed4>] driver_attach+0x11/0x13 Apr 13 15:21:31 bakup kernel: [<c0563048>] __driver_attach+0x0/0x6b Apr 13 15:21:31 bakup kernel: [<c056263a>] bus_add_driver+0x64/0xfd Apr 13 15:21:31 bakup kernel: [<c04febe7>] __pci_register_driver+0x42/0x8c Apr 13 15:21:31 bakup kernel: [<f881d05b>] megaraid_init+0x5b/0x9b [megaraid] Apr 13 15:21:31 bakup kernel: [<c044001f>] sys_init_module+0x1b00/0x1cc5 Apr 13 15:21:31 bakup kernel: [<f8973000>] asus_hotk_remove+0x0/0x173 [asus_acpi] Apr 13 15:21:31 bakup kernel: [<c042aff4>] __request_region+0x0/0xb0 Apr 13 15:21:31 bakup kernel: [<c047866c>] generic_file_llseek+0xc6/0xd2 Apr 13 15:21:31 bakup kernel: [<c0478856>] sys_read+0x3c/0x63 Apr 13 15:21:31 bakup kernel: [<c0404f4b>] syscall_call+0x7/0xb Apr 13 15:21:31 bakup kernel: ======================= Apr 13 15:21:31 bakup kernel: INFO: task scsi_eh_2:1019 blocked for more than 120 seconds. Apr 13 15:21:31 bakup kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. Apr 13 15:21:31 bakup kernel: scsi_eh_2 D 00000011 3472 1019 11 1267 426 (L-TLB) Apr 13 15:21:31 bakup kernel: f7899ee8 00000046 56ce73aa 00000011 00000000 ffffffff ffffffff 0000000a Apr 13 15:21:31 bakup kernel: f78e9aa0 56cefd9c 00000011 000089f2 00000001 f78e9bac c2013544 f7cae3c0 Apr 13 15:21:31 bakup kernel: f78ea7bc 00000000 f7a43008 f78ea30c f7899f58 f78ea7de 00000282 ffffffff Apr 13 15:21:31 bakup kernel: Call Trace: Apr 13 15:21:31 bakup kernel: [<c0623b1d>] wait_for_completion+0x6b/0x8f Apr 13 15:21:31 bakup kernel: [<c041f863>] default_wake_function+0x0/0xc Apr 13 15:21:31 bakup kernel: [<f89aa95e>] mega_internal_command+0xe2/0x146 [megaraid] Apr 13 15:21:31 bakup kernel: [<f89aaa3a>] megaraid_reset+0x28/0x71 [megaraid] Apr 13 15:21:31 bakup kernel: [<f885c206>] __scsi_iterate_devices+0x45/0x58 [scsi_mod] Apr 13 15:21:31 bakup kernel: [<f885e486>] scsi_try_bus_device_reset+0x1a/0x33 [scsi_mod] Apr 13 15:21:31 bakup kernel: [<f885f1f7>] scsi_eh_ready_devs+0x174/0x404 [scsi_mod] Apr 13 15:21:31 bakup kernel: [<f885fbff>] scsi_error_handler+0x2e4/0x43b [scsi_mod] Apr 13 15:21:31 bakup kernel: [<f885f91b>] scsi_error_handler+0x0/0x43b [scsi_mod] Apr 13 15:21:31 bakup kernel: [<c0436ed2>] kthread+0xc0/0xee Apr 13 15:21:31 bakup kernel: [<c0436e12>] kthread+0x0/0xee Apr 13 15:21:31 bakup kernel: [<c0405c87>] kernel_thread_helper+0x7/0x10 Apr 13 15:21:31 bakup kernel: =======================
-----Original Message----- From: centos-bounces@centos.org [mailto:centos-bounces@centos.org] On Behalf Of Dale Dellutri Sent: Friday, April 13, 2012 3:07 PM To: CentOS mailing list Subject: Re: [CentOS] upgrade issue
On Fri, Apr 13, 2012 at 2:02 PM, Paul A razor@meganet.net wrote:
I'm running Raid with scsi disks so I'm assuming it's needed correct?
iSCSI is for carrying SCSI command s and data over IP networks. I don't know how your RAID is set up, but it isn't normally done with iSCSI. See, for example: http://en.wikipedia.org/wiki/ISCSI versus http://en.wikipedia.org/wiki/RAID
-----Original Message----- From: centos-bounces@centos.org [mailto:centos-bounces@centos.org] On Behalf Of Dale Dellutri Sent: Friday, April 13, 2012 1:32 PM To: CentOS mailing list Subject: Re: [CentOS] upgrade issue
On Fri, Apr 13, 2012 at 12:19 PM, Paul A razor@meganet.net wrote:
I have an old dell 2450 that was running kernel 2.6.18-308.1.1.el5 I then upgraded it via yum to 2.6.18-308.1.1.el5. Now when the server boots to the new kernel, 2.6.18-308.1.1.el5 it hangs on "starting iSCSI" and the weird thing is when I tried to switch back to the older 2.6.18-308.1.1.el5 kernel it works fine but the network script fails to start. I'm not really sure what I should do, if someone can give me an idea on what I need to do to fix the iSCSId issue on the new kernel or revert back to the old kernel and fix the network issue. The odd thing is both kernel load the e100 network driver but on the older kernel
I can get the network script to start.
I would appreciate some help.
Do you need iscsi? If not, boot with the old kernel, and disable iscsi
# chkconfig --list | grep iscsi (will probably show iscsi and iscsid)
Then # chkconfig iscsi off # chkconfig iscsid off
Then reboot with the new kernel.
-- Dale Dellutri _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
-- Dale Dellutri _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Paul A wrote:
Dale, I disabled it and the server came up but im seeing all sorts of kernel errors when its disabled ( see below ). If this can't be fixed I would really like to boot the previous kernel however the network script won't start. Im not really sure how upgrading the kernel caused the network to stop working on the previous kernel.
Apr 13 15:21:31 bakup kernel: INFO: task modprobe:740 blocked for more than 120 seconds. Apr 13 15:21:31 bakup kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Oh, wonderful. Didn't used to see that before the last six months.
<snip>
Apr 13 15:21:31 bakup kernel: Call Trace: Apr 13 15:21:31 bakup kernel: [<f89aa522>] megaraid_queue+0x8ce/0x8d8 [megaraid]
<snip> Interesting. Is this built-in for CentOS, or did you install megaraid from other than the CentOS repositories?
mark
Mark, I used the centos repos, this server has centos 5.8.
Iscsi will work on my previous kernel but not on the new one. I wanted to switch back to the previous kernel but for some reason the network scripts hangs even though it shows the e100 nic driver loaded.
-----Original Message----- From: centos-bounces@centos.org [mailto:centos-bounces@centos.org] On Behalf Of m.roth@5-cent.us Sent: Friday, April 13, 2012 3:31 PM To: CentOS mailing list Subject: Re: [CentOS] upgrade issue
Paul A wrote:
Dale, I disabled it and the server came up but im seeing all sorts of kernel errors when its disabled ( see below ). If this can't be fixed I would really like to boot the previous kernel however the network script won't start. Im not really sure how upgrading the kernel caused the network to stop working on the previous kernel.
Apr 13 15:21:31 bakup kernel: INFO: task modprobe:740 blocked for more than 120 seconds. Apr 13 15:21:31 bakup kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Oh, wonderful. Didn't used to see that before the last six months.
<snip>
Apr 13 15:21:31 bakup kernel: Call Trace: Apr 13 15:21:31 bakup kernel: [<f89aa522>] megaraid_queue+0x8ce/0x8d8 [megaraid]
<snip> Interesting. Is this built-in for CentOS, or did you install megaraid from other than the CentOS repositories?
mark
_______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
On Fri, Apr 13, 2012 at 2:25 PM, Paul A razor@meganet.net wrote:
Dale, I disabled it and the server came up but im seeing all sorts of kernel errors when its disabled ( see below ). If this can't be fixed I would really like to boot the previous kernel however the network script won't start. Im not really sure how upgrading the kernel caused the network to stop working on the previous kernel.
Nor do I know how upgrading could do that. What other changes have you made?
Two questions: 1. Is the previous kernel (with iscsi and iscsid enabled) giving any error messages in the log when you attempt to boot it. 2. Is your RAID using local disks or disks on an iscsi target device? If local, iscsi shouldn't have anything to do with RAID. If on an iscsi target device, then you need iscsi and iscsid enabled.
I'd suggest re-enabling iscsi and iscsid and try to solve the network problem with the old kernel.
-----Original Message----- From: centos-bounces@centos.org [mailto:centos-bounces@centos.org] On Behalf Of Dale Dellutri Sent: Friday, April 13, 2012 3:07 PM To: CentOS mailing list Subject: Re: [CentOS] upgrade issue
On Fri, Apr 13, 2012 at 2:02 PM, Paul A razor@meganet.net wrote:
I'm running Raid with scsi disks so I'm assuming it's needed correct?
iSCSI is for carrying SCSI command s and data over IP networks. I don't know how your RAID is set up, but it isn't normally done with iSCSI. See, for example: http://en.wikipedia.org/wiki/ISCSI versus http://en.wikipedia.org/wiki/RAID
-----Original Message----- From: centos-bounces@centos.org [mailto:centos-bounces@centos.org] On Behalf Of Dale Dellutri Sent: Friday, April 13, 2012 1:32 PM To: CentOS mailing list Subject: Re: [CentOS] upgrade issue
On Fri, Apr 13, 2012 at 12:19 PM, Paul A razor@meganet.net wrote:
I have an old dell 2450 that was running kernel 2.6.18-308.1.1.el5 I then upgraded it via yum to 2.6.18-308.1.1.el5. Now when the server boots to the new kernel, 2.6.18-308.1.1.el5 it hangs on "starting iSCSI" and the weird thing is when I tried to switch back to the older 2.6.18-308.1.1.el5 kernel it works fine but the network script fails to start. I'm not really sure what I should do, if someone can give me an idea on what I need to do to fix the iSCSId issue on the new kernel or revert back to the old kernel and fix the network issue. The odd thing is both kernel load the e100 network driver but on the older kernel
I can get the network script to start.
I would appreciate some help.
Do you need iscsi? If not, boot with the old kernel, and disable iscsi
# chkconfig --list | grep iscsi (will probably show iscsi and iscsid)
Then # chkconfig iscsi off # chkconfig iscsid off
Then reboot with the new kernel.
-- Dale Dellutri _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
-- Dale Dellutri _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
-
Dale Dellutri -
m.roth＠5-cent.us -
Paul A