FAQ Search Today's Posts Mark Forums Read
» Video Reviews

» Linux Archive

Linux-archive is a website aiming to archive linux email lists and to make them easily accessible for linux users/developers.


» Sponsor

» Partners

» Sponsor

Go Back   Linux Archive > Redhat > Device-mapper Development

 
 
LinkBack Thread Tools
 
Old 11-28-2011, 04:01 PM
"Moger, Babu"
 
Default info on enabling only one path with rdac and DS4700

> -----Original Message-----
> From: Gianluca Cecchi [mailto:gianluca.cecchi@gmail.com]
> Sent: Monday, November 28, 2011 10:05 AM
> To: device-mapper development
> Subject: Re: [dm-devel] info on enabling only one path with rdac and
> DS4700
>
> On Mon, Nov 28, 2011 at 3:46 PM, Moger, Babu wrote:
> >> 3) Based on messages during failover, is it true that I can avoid
> >> explicitly put scsi_dh in initrd?
> >> If I create initrd this way:
> >> mkinitrd /boot/initrd-$(uname -r)-scsi_dh.img $(uname -r) --
> >> preload=scsi_dh_rdac
> >> I get this difference:
> >> [root@testserver ~]# diff /tmp/new/init /tmp/current/init
> >> 44,51d43
> >> < echo "Loading scsi_mod.ko module"
> >> < insmod /lib/scsi_mod.ko
> >> < echo "Loading sd_mod.ko module"
> >> < insmod /lib/sd_mod.ko
> >> < echo "Loading scsi_dh.ko module"
> >> < insmod /lib/scsi_dh.ko
> >> < echo "Loading scsi_dh_rdac.ko module"
> >> < insmod /lib/scsi_dh_rdac.ko
> >> 62a55,58
> >> > echo "Loading scsi_mod.ko module"
> >> > insmod /lib/scsi_mod.ko
> >> > echo "Loading sd_mod.ko module"
> >> > insmod /lib/sd_mod.ko
> >>
> >> or will it help in any way?
> >
> > Having scsi_dh_rdac in initrd will help to get rid of the initial I/O
> errors you are seeing.
> >
>
> Actually not.... see screen shots of console start at
> https://docs.google.com/open?id=0BwoPbcrMv8mvMWRlYmVhM2MtY2JkYi00ZDMzLW
> IxMWMtMDM0YzZjMWU0NGE1
> and
> https://docs.google.com/open?id=0BwoPbcrMv8mvM2RjMDg1MWUtYzVmZC00OTA1LT
> ljNjAtZGZmMjM5NDlhYTI4
> and also dmesg giving for example for sdb and sdc:

Looks like there is something else going on here.

>
> scsi 3:0:0:1: rdac: LUN 1 (unowned)

This is when handler is loaded. Looks like your passive paths were detected first.
>From now on, I/Os coming on to sdb will be silenced and will not got to target. That is not happening.

What is the kernel level. I remember there was an issue. It was fixed later by this commit..
http://git.kernel.org/?p=linux/kernel/git/next/linux-next.git;a=commitdiff;h=5a36756b8b3d7178d9c101d1e2 a7a83fd7b17706



> sdb : very big device. try to use READ CAPACITY(16).
> SCSI device sdb: 7320494080 512-byte hdwr sectors (3748093 MB)
> sdb: Write Protect is off
> sdb: Mode Sense: 77 00 10 08
> SCSI device sdb: drive cache: write back w/ FUA
> sdb : very big device. try to use READ CAPACITY(16).
> SCSI device sdb: 7320494080 512-byte hdwr sectors (3748093 MB)
> sdb: Write Protect is off
> sdb: Mode Sense: 77 00 10 08
> SCSI device sdb: drive cache: write back w/ FUA
> sdb:<3>Buffer I/O error on device sdb, logical block 0
> Buffer I/O error on device sdb, logical block 0
> Buffer I/O error on device sdb, logical block 0
> Buffer I/O error on device sdb, logical block 0
> Buffer I/O error on device sdb, logical block 0
> Buffer I/O error on device sdb, logical block 0
> Buffer I/O error on device sdb, logical block 0
> Dev sdb: unable to read RDB block 0
> Buffer I/O error on device sdb, logical block 0
> Buffer I/O error on device sdb, logical block 0
> unable to read partition table
> sd 3:0:0:1: Attached scsi disk sdb
> Vendor: IBM Model: 1814 FAStT Rev: 0916
> Type: Direct-Access ANSI SCSI revision: 05
> scsi 3:0:1:1: rdac: LUN 1 (owned)
> scsi 3:0:1:1: rdac: LUN 1 (owned)
> sdc : very big device. try to use READ CAPACITY(16).
> SCSI device sdc: 7320494080 512-byte hdwr sectors (3748093 MB)
> sdc: Write Protect is off
> sdc: Mode Sense: 77 00 10 08
> SCSI device sdc: drive cache: write back w/ FUA
> sdc : very big device. try to use READ CAPACITY(16).
> SCSI device sdc: 7320494080 512-byte hdwr sectors (3748093 MB)
> sdc: Write Protect is off
> sdc: Mode Sense: 77 00 10 08
> SCSI device sdc: drive cache: write back w/ FUA
> sdc: unknown partition table
> sd 3:0:1:1: Attached scsi disk sdc
> ...
> GSI 23 sharing vector 0x72 and IRQ 23
> ACPI: PCI Interrupt 0000:04:00.0[A] -> GSI 17 (level, low) -> IRQ 114
> eth0: Broadcom NetXtreme II BCM5708 1000Base-SX (B2) PCI-X 64-bit
> 133MHz found at mem da000000, IRQ 114, node addr 00215e2fd0e2
> ACPI: PCI Interrupt 0000:06:00.0[A] -> GSI 19 (level, low) -> IRQ 90
> eth1: Broadcom NetXtreme II BCM5708 1000Base-SX (B2) PCI-X 64-bit
> 133MHz found at mem d8000000, IRQ 90, node addr 00215e2fd0e4
> printk: 69 messages suppressed.
> Buffer I/O error on device sdb, logical block 0
> ...
> md: Autodetecting RAID arrays.
> md: autorun ...
> md: ... autorun DONE.
> device-mapper: multipath: version 1.0.6 loaded
> device-mapper: multipath: Using scsi_dh module scsi_dh_rdac for
> failover/failback and device management.
> device-mapper: multipath round-robin: version 1.0.0 loaded
> printk: 4 messages suppressed.
> Buffer I/O error on device sdb, logical block 915061759
> ...
>
> So if I have to get yet the I/O errors, having no advantage, I'd
> prefer to stay stick with the default initrd...
>
> --
> dm-devel mailing list
> dm-devel@redhat.com
> https://www.redhat.com/mailman/listinfo/dm-devel

--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel
 
Old 11-28-2011, 04:26 PM
Gianluca Cecchi
 
Default info on enabling only one path with rdac and DS4700

On Mon, Nov 28, 2011 at 6:01 PM, Moger, Babu wrote:
>
> Looks like there is something else going on here.
>
>>
>> scsi 3:0:0:1: rdac: LUN 1 (unowned)
>
> This is when handler is loaded. *Looks like your passive paths were detected first.
> >From now on, I/Os coming on to sdb will be silenced and will not got to target. That is not happening.
>
> What is the kernel level. I remember there was an issue. It was fixed later by this commit..
> http://git.kernel.org/?p=linux/kernel/git/next/linux-next.git;a=commitdiff;h=5a36756b8b3d7178d9c101d1e2 a7a83fd7b17706


the kernel is latest from RH EL 5.7 + updates:
2.6.18-274.7.1.el5

I'm going to check with its kernel-source to see it the patch is
applied or not...

Gianluca

--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel
 
Old 11-28-2011, 05:05 PM
"Moger, Babu"
 
Default info on enabling only one path with rdac and DS4700

> -----Original Message-----
> From: Gianluca Cecchi [mailto:gianluca.cecchi@gmail.com]
> Sent: Monday, November 28, 2011 11:26 AM
> To: device-mapper development
> Subject: Re: [dm-devel] info on enabling only one path with rdac and
> DS4700
>
> On Mon, Nov 28, 2011 at 6:01 PM, Moger, Babu wrote:
> >
> > Looks like there is something else going on here.
> >
> >>
> >> scsi 3:0:0:1: rdac: LUN 1 (unowned)
> >
> > This is when handler is loaded. *Looks like your passive paths were
> detected first.
> > >From now on, I/Os coming on to sdb will be silenced and will not got
> to target. That is not happening.
> >
> > What is the kernel level. I remember there was an issue. It was fixed
> later by this commit..
> > http://git.kernel.org/?p=linux/kernel/git/next/linux-
> next.git;a=commitdiff;h=5a36756b8b3d7178d9c101d1e2 a7a83fd7b17706
>
>
> the kernel is latest from RH EL 5.7 + updates:
> 2.6.18-274.7.1.el5
>
> I'm going to check with its kernel-source to see it the patch is
> applied or not...

RHEL 5.7 has the patch already as I know. You can check to be sure. Can you also add "scsi_mod.scsi_logging_level=5056" in kernel line to get more information on what else is going on.

>
> Gianluca
>
> --
> dm-devel mailing list
> dm-devel@redhat.com
> https://www.redhat.com/mailman/listinfo/dm-devel

--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel
 
Old 11-28-2011, 07:13 PM
Gianluca Cecchi
 
Default info on enabling only one path with rdac and DS4700

On Mon, Nov 28, 2011 at 7:05 PM, Moger, Babu wrote:

>>
>>
>> the kernel is latest from RH EL 5.7 + updates:
>> 2.6.18-274.7.1.el5
>>
>> I'm going to check with its kernel-source to see it the patch is
>> applied or not...
>
> RHEL 5.7 has the patch already as I know. You can check to be sure. Can you also add "scsi_mod.scsi_logging_level=5056" in kernel line to get more information on what else is going on.

Ok, I have verified that the patch should be applied.
Downloaded kernel-2.6.18-274.7.1.el5.src.rpm
and in spec file I can find referral to
kernel-2.6.18-redhat.patch
that contains this inside section
--- linux-2.6.18.4/drivers/scsi/device_handler/scsi_dh_rdac.c
+++ linux-2.6.18-redhat/drivers/scsi/device_handler/scsi_dh_rdac.c
...
+static int check_ownership(struct scsi_device *sdev, struct rdac_dh_data *h)
+{
+ int err;
+ struct c9_inquiry *inqp;
+
+ h->lun_state = RDAC_LUN_UNOWNED;
+ h->state = RDAC_STATE_ACTIVE;
+ err = submit_inquiry(sdev, 0xC9, sizeof(struct c9_inquiry), h);
+ if (err == SCSI_DH_OK) {
+ inqp = &h->inq.c9;
+ if ((inqp->avte_cvp >> 7) == 0x1) {
+ /* LUN in AVT mode */
+ sdev_printk(KERN_NOTICE, sdev,
+ "%s: AVT mode detected
",
+ RDAC_NAME);
+ h->lun_state = RDAC_LUN_AVT;
+ } else if ((inqp->avte_cvp & 0x1) != 0) {
+ /* LUN was owned by the controller */
+ h->lun_state = RDAC_LUN_OWNED;
+ }
+ }
+
+ if (h->lun_state == RDAC_LUN_UNOWNED)
+ h->state = RDAC_STATE_PASSIVE;
+ return err;
+}
+

So I'm going to add the scsi logging parameter you suggested and see.

--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel
 
Old 11-29-2011, 05:29 AM
Gianluca Cecchi
 
Default info on enabling only one path with rdac and DS4700

On Mon, Nov 28, 2011 at 9:13 PM, Gianluca Cecchi wrote:
> On Mon, Nov 28, 2011 at 7:05 PM, Moger, Babu wrote:
[snip]
>> RHEL 5.7 has the patch already as I know. You can check to be sure. Can you also add "scsi_mod.scsi_logging_level=5056" in kernel line to get more information on what else is going on.
>
> Ok, I have verified that the patch should be applied.
[snip]
> So I'm going to add the scsi logging parameter you suggested and see.
>

Unfortunately it seems that rh el 5.x kernel doesn't support that
scsi_mod kernel module option directly at kernel line
# cat /proc/cmdline
ro root=/dev/VolGroup00/LV_root rhgb quiet scsi_mod.scsi_logging_level=5056

In messages:
Nov 29 07:00:03 testserver kernel: Unknown boot option
`scsi_mod.scsi_logging_level=5056': ignoring

So I picked up the new initrd with the scsi_dh part and put directly
that option at insmod statement line for scsi_mod module in initrd
init file, creating another special grub.conf stanza for this initrd
See messages and output of dmesg with this parameter at the links below

https://docs.google.com/open?id=0BwoPbcrMv8mvNzU1ZTNlY2UtY2E5Ni00NDdlLWIyY 2UtZmQ0Y2I4OTJmNTZi
and
https://docs.google.com/open?id=0BwoPbcrMv8mvZmUxNTNmNjgtOTZmMi00YTBiLTkyZ TktZWU0NTdjZWI4OWZm

At the end of boot:
mpath1 (3600a0b80005012440000093e4a55cf33) dm-6 IBM,1814 FAStT
[size=3.4T][features=1 queue_if_no_path][hwhandler=1 rdac][rw]
\_ round-robin 0 [prio=0][active]
\_ 3:0:1:1 sdc 8:32 [active][undef]
\_ 4:0:1:1 sde 8:64 [active][undef]
\_ round-robin 0 [prio=0][enabled]
\_ 3:0:0:1 sdb 8:16 [active][undef]
\_ 4:0:0:1 sdd 8:48 [active][undef]

Hope it helps you understanding better what doesn't go as expected..
Thanks for your time

Gianluca

--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel
 
Old 11-29-2011, 04:07 PM
"Moger, Babu"
 
Default info on enabling only one path with rdac and DS4700

I think I got it what is happening.. Please see my comments below..

> -----Original Message-----
> From: Gianluca Cecchi [mailto:gianluca.cecchi@gmail.com]
> Sent: Tuesday, November 29, 2011 12:29 AM
> To: device-mapper development
> Subject: Re: [dm-devel] info on enabling only one path with rdac and
> DS4700
>
> On Mon, Nov 28, 2011 at 9:13 PM, Gianluca Cecchi wrote:
> > On Mon, Nov 28, 2011 at 7:05 PM, Moger, Babu wrote:
> [snip]
> >> RHEL 5.7 has the patch already as I know. You can check to be sure.
> Can you also add "scsi_mod.scsi_logging_level=5056" in kernel line to
> get more information on what else is going on.
> >
> > Ok, I have verified that the patch should be applied.
> [snip]
> > So I'm going to add the scsi logging parameter you suggested and see.
> >
>
> Unfortunately it seems that rh el 5.x kernel doesn't support that
> scsi_mod kernel module option directly at kernel line
> # cat /proc/cmdline
> ro root=/dev/VolGroup00/LV_root rhgb quiet
> scsi_mod.scsi_logging_level=5056
>
> In messages:
> Nov 29 07:00:03 testserver kernel: Unknown boot option
> `scsi_mod.scsi_logging_level=5056': ignoring
>
> So I picked up the new initrd with the scsi_dh part and put directly
> that option at insmod statement line for scsi_mod module in initrd
> init file, creating another special grub.conf stanza for this initrd
> See messages and output of dmesg with this parameter at the links below
>
> https://docs.google.com/open?id=0BwoPbcrMv8mvNzU1ZTNlY2UtY2E5Ni00NDdlLW
> IyY2UtZmQ0Y2I4OTJmNTZi
> and
> https://docs.google.com/open?id=0BwoPbcrMv8mvZmUxNTNmNjgtOTZmMi00YTBiLT
> kyZTktZWU0NTdjZWI4OWZm
>
> At the end of boot:
> mpath1 (3600a0b80005012440000093e4a55cf33) dm-6 IBM,1814 FAStT
> [size=3.4T][features=1 queue_if_no_path][hwhandler=1 rdac][rw]
> \_ round-robin 0 [prio=0][active]
> \_ 3:0:1:1 sdc 8:32 [active][undef]
> \_ 4:0:1:1 sde 8:64 [active][undef]
> \_ round-robin 0 [prio=0][enabled]
> \_ 3:0:0:1 sdb 8:16 [active][undef]
> \_ 4:0:0:1 sdd 8:48 [active][undef]
>
> Hope it helps you understanding better what doesn't go as expected..
> Thanks for your time

Here are the messages without scsi_dh_rdac in initrd.
Nov 23 17:32:56 testserver kernel: sdc:end_request: I/O error, dev
sdb, sector 0
Nov 23 17:32:56 testserver kernel: Buffer I/O error on device sdb,
logical block 0

Here are the messages with scsi_dh_rdac in initrd.

Nov 29 07:16:46 testserver kernel: Buffer I/O error on device sdd, logical block 0

If you notice above, there is only one message in having driver in initrd. The messages " Buffer I/O error" is coming from file system layer. There is no way of silencing there errors in older kernels. These messages are harmless. The newer kernels use "BH_Quiet" flag to silence there error. You can check the newer kernel sources.

Final note. It is always recommended to include the handlers in initrd because it will avoid I/Os to passive paths reducing the boot time.



>
> Gianluca
>
> --
> dm-devel mailing list
> dm-devel@redhat.com
> https://www.redhat.com/mailman/listinfo/dm-devel

--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel
 

Thread Tools




All times are GMT. The time now is 01:15 AM.

VBulletin, Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO ©2007, Crawlability, Inc.
Copyright 2007 - 2008, www.linux-archive.org