RE: [PATCH 4/4] [libmultipath] Update hardware table for HPHSV200

RE: [PATCH 4/4] [libmultipath] Update hardware table for HPHSV200

[Reilly, Stephen (MRO)]

The product name does change if the firmware uses 'START_STOP_UNIT' or
ALUA... so why are we adding the revision?

-----Original Message-----
From: dm-devel-bounces@redhat.com [mailto:dm-devel-bounces@redhat.com]
On Behalf Of Hannes Reinecke
Sent: Thursday, November 09, 2006 5:10 AM
Subject: [dm-devel] [PATCH 4/4] [libmultipath] Update hardware table for
HPHSV200

According to HP the behaviour of the HSV200 depends on the software
release; all revisions smaller or equal to '3000' have active/passive
support; all revisions above that have active/active support.
Translated it means that the older firmware uses the HP-style
'START_STOP_UNIT' command to switch paths, while the newer firmware
supports ALUA.
So we should update the hardware tables accordingly.

Signed-off-by: Hannes Reinecke <hare@suse.de>

---

 libmultipath/hwtable.c |   49
+++++++++++++++++++++++++++++++++++++++++++++++-
 1 files changed, 48 insertions(+), 1 deletions(-)

19c1285b869bfe42b3605c2fee73bb4b416c4c40
diff --git a/libmultipath/hwtable.c b/libmultipath/hwtable.c
index b56c231..8c8699d 100644
--- a/libmultipath/hwtable.c
+++ b/libmultipath/hwtable.c
@@ -91,7 +91,7 @@ static struct hwentry default_hw[] = {
 	},
 	{
 		.vendor        = "HP",
-		.product       = "(HSV2.*|A6189A)",
+		.product       = "A6189A",
 		.getuid        = DEFAULT_GETUID,
 		.getprio       = NULL,
 		.features      = DEFAULT_FEATURES,
@@ -104,6 +104,53 @@ static struct hwentry default_hw[] = {
 		.minio         = DEFAULT_MINIO,
 		.checker_name  = READSECTOR0,
 	},
+	{
+		.vendor        = "HP",
+		.product       = "HSV20.*",
+		.revision      = "[123].*",
+		.getuid        = DEFAULT_GETUID,
+		.getprio       = NULL,
+		.features      = DEFAULT_FEATURES,
+		.hwhandler     = DEFAULT_HWHANDLER,
+		.selector      = DEFAULT_SELECTOR,
+		.pgpolicy      = MULTIBUS,
+		.pgfailback    = FAILBACK_UNDEF,
+		.rr_weight     = RR_WEIGHT_NONE,
+		.no_path_retry = NO_PATH_RETRY_UNDEF,
+		.minio         = DEFAULT_MINIO,
+		.checker_name  = HP_SW,
+	},
+	{
+		.vendor        = "HP",
+		.product       = "HSV20.*",
+		.revision      = "[^123].*",
+		.getuid        = DEFAULT_GETUID,
+		.getprio       = "/sbin/mpath_prio_alua /dev/%n",
+		.features      = DEFAULT_FEATURES,
+		.hwhandler     = DEFAULT_HWHANDLER,
+		.selector      = DEFAULT_SELECTOR,
+		.pgpolicy      = MULTIBUS,
+		.pgfailback    = -FAILBACK_IMMEDIATE,
+		.rr_weight     = RR_WEIGHT_PRIORITIES,
+		.no_path_retry = NO_PATH_RETRY_UNDEF,
+		.minio         = DEFAULT_MINIO,
+		.checker_name  = TUR,
+	},
+	{
+		.vendor        = "HP",
+		.product       = "HSV21.*",
+		.getuid        = DEFAULT_GETUID,
+		.getprio       = "/sbin/mpath_prio_alua /dev/%n",
+		.features      = DEFAULT_FEATURES,
+		.hwhandler     = DEFAULT_HWHANDLER,
+		.selector      = DEFAULT_SELECTOR,
+		.pgpolicy      = GROUP_BY_PRIO,
+		.pgfailback    = FAILBACK_UNDEF,
+		.rr_weight     = RR_WEIGHT_NONE,
+		.no_path_retry = NO_PATH_RETRY_UNDEF,
+		.minio         = DEFAULT_MINIO,
+		.checker_name  = TUR,
+	},
 	/*
 	 * DDN controller family
 	 *
-- 
1.3.1


--------------040805080108070409060801--

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

Re: [PATCH 4/4] [libmultipath] Update hardware table for HPHSV200

[Hannes Reinecke]

Reilly, Stephen (MRO) wrote:
> The product name does change if the firmware uses 'START_STOP_UNIT' or
> ALUA... so why are we adding the revision?
> 
Ah, that's news to me.
How does the product name change?
My contacts at HP Boeblingen said that the revision number indicated the 
behaviour.
What are the names of the fw using START_STOP_UNIT?
And what are those for using ALUA?

And irrespective of this I think matching by revision is still a valid 
addition to the hardware table mechanism.

Cheers,

Hannes
-- 
Dr. Hannes Reinecke			hare@suse.de
SuSE Linux Products GmbH		S390 & zSeries
Maxfeldstraße 5				+49 911 74053 688
90409 Nürnberg				http://www.suse.de

RE: [PATCH 4/4] [libmultipath] Update hardware table forHPHSV200

[Reilly, Stephen (MRO)]

-----Original Message-----
From: dm-devel-bounces@redhat.com [mailto:dm-devel-bounces@redhat.com] On Behalf Of Hannes Reinecke
Sent: Tuesday, November 14, 2006 2:21 AM
To: device-mapper development
Subject: Re: [dm-devel] [PATCH 4/4] [libmultipath] Update hardware table forHPHSV200

Reilly, Stephen (MRO) wrote:
> The product name does change if the firmware uses 'START_STOP_UNIT' or
> ALUA... so why are we adding the revision?
> 
Ah, that's news to me.
How does the product name change?
>> The firmware will change the name
My contacts at HP Boeblingen said that the revision number indicated the 
behaviour.
What are the names of the fw using START_STOP_UNIT?
>> HSV100 and HSV110 EVA3000/5000
And what are those for using ALUA?
>> HSV101, HSV111, HSV200 and HSV210 EVA 3000/5000  EVA4000/6000/8000

And irrespective of this I think matching by revision is still a valid 
addition to the hardware table mechanism.
>> However I would not use it determine the START_STOP_UNIT or ALUA implementation

Cheers,

Hannes
-- 
Dr. Hannes Reinecke			hare@suse.de
SuSE Linux Products GmbH		S390 & zSeries
Maxfeldstraße 5				+49 911 74053 688
90409 Nürnberg				http://www.suse.de

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

Re: [PATCH 4/4] [libmultipath] Update hardware table forHPHSV200

[Hannes Reinecke]

[-- Attachment #1: Type: text/plain, Size: 1468 bytes --]

Reilly, Stephen (MRO) wrote:
[ Quoting appears to be a lost art ... ]
> 
> Hannes Reinecke wrote:
 >> Reilly, Stephen (MRO) wrote:
>>> The product name does change if the firmware uses 'START_STOP_UNIT' or
>>> ALUA... so why are we adding the revision?
>>
>> Ah, that's news to me.
>> How does the product name change?
 >
> The firmware will change the name
 >
Oh, that's nice.

>> My contacts at HP Boeblingen said that the revision number indicated the 
>> behaviour.
>> What are the names of the fw using START_STOP_UNIT?
> HSV100 and HSV110 EVA3000/5000
EVA is the name of the machine; HSV is the SCSI model name, right?

>> And what are those for using ALUA?
> HSV101, HSV111, HSV200 and HSV210 EVA 3000/5000  EVA4000/6000/8000
> 
What about the old MSA ones? Is there a firmware upgrade available?
If so, will the name change there, too?

>> And irrespective of this I think matching by revision is still a valid 
>> addition to the hardware table mechanism.
> However I would not use it determine the START_STOP_UNIT or ALUA implementation
> 
I was under the impression that the name would stay the same. If that's 
not the case then of course we should match onto the name.

Please check the attached patch if it updates the hwtables correctly.

Cheers,

Hannes
-- 
Dr. Hannes Reinecke			hare@suse.de
SuSE Linux Products GmbH		S390 & zSeries
Maxfeldstraße 5				+49 911 74053 688
90409 Nürnberg				http://www.suse.de

[-- Attachment #2: hp-hwtable-update --]
[-- Type: text/plain, Size: 2961 bytes --]

diff --git a/libmultipath/hwtable.c b/libmultipath/hwtable.c
index ac126c0..2fd6fac 100644
--- a/libmultipath/hwtable.c
+++ b/libmultipath/hwtable.c
@@ -75,21 +75,6 @@ static struct hwentry default_hw[] = {
 		.checker_name  = HP_SW,
 	},
 	{
-		.vendor        = "(COMPAQ|HP)",
-		.product       = "(MSA|HSV)1.*",
-		.getuid        = DEFAULT_GETUID,
-		.getprio       = NULL,
-		.features      = DEFAULT_FEATURES,
-		.hwhandler     = "1 hp_sw",
-		.selector      = DEFAULT_SELECTOR,
-		.pgpolicy      = MULTIBUS,
-		.pgfailback    = FAILBACK_UNDEF,
-		.rr_weight     = RR_WEIGHT_NONE,
-		.no_path_retry = NO_PATH_RETRY_UNDEF,
-		.minio         = DEFAULT_MINIO,
-		.checker_name  = HP_SW,
-	},
-	{
 		.vendor        = "HP",
 		.product       = "A6189A",
 		.getuid        = DEFAULT_GETUID,
@@ -105,13 +90,13 @@ static struct hwentry default_hw[] = {
 		.checker_name  = READSECTOR0,
 	},
 	{
-		.vendor        = "HP",
-		.product       = "HSV20.*",
-		.revision      = "[123].*",
+		/* EVA 3000/5000 with old firmware */
+		.vendor        = "(COMPAQ|HP)",
+		.product       = "(MSA|HSV)1.0.*",
 		.getuid        = DEFAULT_GETUID,
 		.getprio       = NULL,
 		.features      = DEFAULT_FEATURES,
-		.hwhandler     = DEFAULT_HWHANDLER,
+		.hwhandler     = "1 hp_sw",
 		.selector      = DEFAULT_SELECTOR,
 		.pgpolicy      = MULTIBUS,
 		.pgfailback    = FAILBACK_UNDEF,
@@ -121,15 +106,15 @@ static struct hwentry default_hw[] = {
 		.checker_name  = HP_SW,
 	},
 	{
-		.vendor        = "HP",
-		.product       = "HSV20.*",
-		.revision      = "[^123].*",
+		/* EVA 3000/5000 with new firmware */
+		.vendor        = "(COMPAQ|HP)",
+		.product       = "(MSA|HSV)1.1.*",
 		.getuid        = DEFAULT_GETUID,
 		.getprio       = "/sbin/mpath_prio_alua /dev/%n",
 		.features      = DEFAULT_FEATURES,
 		.hwhandler     = DEFAULT_HWHANDLER,
 		.selector      = DEFAULT_SELECTOR,
-		.pgpolicy      = MULTIBUS,
+		.pgpolicy      = GROUP_BY_PRIO,
 		.pgfailback    = -FAILBACK_IMMEDIATE,
 		.rr_weight     = RR_WEIGHT_NONE,
 		.no_path_retry = NO_PATH_RETRY_UNDEF,
@@ -137,15 +122,16 @@ static struct hwentry default_hw[] = {
 		.checker_name  = TUR,
 	},
 	{
+		/* EVA 4000/6000/8000 */
 		.vendor        = "HP",
-		.product       = "HSV21.*",
+		.product       = "HSV2.*",
 		.getuid        = DEFAULT_GETUID,
 		.getprio       = "/sbin/mpath_prio_alua /dev/%n",
 		.features      = DEFAULT_FEATURES,
 		.hwhandler     = DEFAULT_HWHANDLER,
 		.selector      = DEFAULT_SELECTOR,
 		.pgpolicy      = GROUP_BY_PRIO,
-		.pgfailback    = FAILBACK_UNDEF,
+		.pgfailback    = -FAILBACK_IMMEDIATE,
 		.rr_weight     = RR_WEIGHT_NONE,
 		.no_path_retry = NO_PATH_RETRY_UNDEF,
 		.minio         = DEFAULT_MINIO,
@@ -249,7 +235,7 @@ static struct hwentry default_hw[] = {
 		.rr_weight     = RR_WEIGHT_NONE,
 		.no_path_retry = NO_PATH_RETRY_UNDEF,
 		.minio         = DEFAULT_MINIO,
-		.checker_name  = READSECTOR0,
+		.checker_name  = TUR,
 	},
 	{
 		.vendor        = "HITACHI",

[-- Attachment #3: Type: text/plain, Size: 0 bytes --]

RE: [PATCH 4/4] [libmultipath] Update hardware table forHPHSV200

[Reilly, Stephen (MRO)]

-----Original Message-----
From: Hannes Reinecke [mailto:hare@suse.de] 
Sent: Wednesday, November 15, 2006 7:50 AM
To: Reilly, Stephen (MRO)
Cc: device-mapper development
Subject: Re: [dm-devel] [PATCH 4/4] [libmultipath] Update hardware table forHPHSV200

Reilly, Stephen (MRO) wrote:
[ Quoting appears to be a lost art ... ]
> 
> Hannes Reinecke wrote:
 >> Reilly, Stephen (MRO) wrote:
>>> The product name does change if the firmware uses 'START_STOP_UNIT' or
>>> ALUA... so why are we adding the revision?
>>
>> Ah, that's news to me.
>> How does the product name change?
 >
> The firmware will change the name
 >
Oh, that's nice.

>> My contacts at HP Boeblingen said that the revision number indicated the 
>> behaviour.
>> What are the names of the fw using START_STOP_UNIT?
> HSV100 and HSV110 EVA3000/5000
EVA is the name of the machine; HSV is the SCSI model name, right?
[Reilly, Stephen (MRO)] Correct, presented both for reference

>> And what are those for using ALUA?
> HSV101, HSV111, HSV200 and HSV210 EVA 3000/5000  EVA4000/6000/8000
> 
What about the old MSA ones? Is there a firmware upgrade available?
If so, will the name change there, too?

			 Vendor Id: 		Product Id: 
MSA1500 A-A -      "HP      "		      "MSA VOLUME      "

MSA1000 A-P - 	 "COMPAQ  "			"MSA1000 VOLUME  "
MSA1500 A-P -      "COMPAQ  "			"MSA1000 VOLUME  "


>> And irrespective of this I think matching by revision is still a valid 
>> addition to the hardware table mechanism.
> However I would not use it determine the START_STOP_UNIT or ALUA implementation
> 
I was under the impression that the name would stay the same. If that's 
not the case then of course we should match onto the name.

Please check the attached patch if it updates the hwtables correctly.

Cheers,

Hannes
-- 
Dr. Hannes Reinecke			hare@suse.de
SuSE Linux Products GmbH		S390 & zSeries
Maxfeldstraße 5				+49 911 74053 688
90409 Nürnberg				http://www.suse.de