multipathd: sdc: readsector0 checker reports path is down

multipathd: sdc: readsector0 checker reports path is down

[Ray Van Dolson]

I'm still getting the hang of iSCSI and multipath, so bear with me if
this is a FAQ that I've missed...

I have a host attaching to an MD3000i via iSCSI/dm-multipath that is
working but showing a lot of the following errors in syslog:

  multipathd: sdc: readsector0 checker reports path is down

My initial impression is that sdc is likely the "passive" path (for
failover) whereas sdd is my active path:

  # multipath -ll
  sdc: checker msg is "readsector0 checker reports path is down"
  mpath2 (36001ec9000f338370000000000000000) dm-2 DELL,Universal Xport
  [size=20M][features=0][hwhandler=0]
  \_ round-robin 0 [prio=0][enabled]
   \_ 4:0:0:31 sdf 8:80  [active][ready] 
  mpath1 (36001ec9000f337dd0000000000000000) dm-1 DELL,Universal Xport
  [size=20M][features=0][hwhandler=0]
  \_ round-robin 0 [prio=0][enabled]
   \_ 3:0:0:31 sde 8:64  [active][ready] 
  mpath0 (36001ec9000f337dd0000065b491854a4) dm-0 DELL,MD3000i
  [size=150G][features=0][hwhandler=1 rdac]
  \_ round-robin 0 [prio=3][active]
   \_ 3:0:0:0  sdd 8:48  [active][ready] 
  \_ round-robin 0 [prio=0][enabled]
   \_ 4:0:0:0  sdc 8:32  [failed][faulty]

  # iscsiadm -m discovery -t sendtargets -p 192.168.132.101:3260
  0.0.0.0:3260,1 iqn.1984-05.com.dell:powervault.6001ec9000f337dd0000000048bdc862
  0.0.0.0:3260,2 iqn.1984-05.com.dell:powervault.6001ec9000f337dd0000000048bdc862
  192.168.132.101:3260,1 iqn.1984-05.com.dell:powervault.6001ec9000f337dd0000000048bdc862
  192.168.132.102:3260,2 iqn.1984-05.com.dell:powervault.6001ec9000f337dd0000000048bdc862

  # iscsiadm -m session -P 1
  Target: iqn.1984-05.com.dell:powervault.6001ec9000f337dd0000000048bdc862
    Current Portal: 192.168.132.101:3260,1
    Persistent Portal: 192.168.132.101:3260,1
      **********
      Interface:
      **********
      Iface Name: default
      Iface Transport: tcp
      Iface Initiatorname: iqn.1994-05.com.redhat:2738e0d3bd21
      Iface IPaddress: 192.168.132.202
      Iface HWaddress: default
      Iface Netdev: default
      SID: 1
      iSCSI Connection State: LOGGED IN
      iSCSI Session State: Unknown
      Internal iscsid Session State: NO CHANGE
    Current Portal: 192.168.132.102:3260,2
    Persistent Portal: 192.168.132.102:3260,2
      **********
      Interface:
      **********
      Iface Name: default
      Iface Transport: tcp
      Iface Initiatorname: iqn.1994-05.com.redhat:2738e0d3bd21
      Iface IPaddress: 192.168.132.202
      Iface HWaddress: default
      Iface Netdev: default
      SID: 2
      iSCSI Connection State: LOGGED IN
      iSCSI Session State: Unknown
      Internal iscsid Session State: NO CHANGE

sde and sdf are the silly 20MB partition that the MD3000 creates and
I'm not using either.

My ultimate goal is to not have so many of the path is down errors
showing up in my syslog file (hundreds and hundreds of them).  I'm not
sure if their presence is normal (indicative of a passive path) or
indicative of a problem.

Also, I'm not clear as to how I should identify which target
corresponds with which drive (sdc or sdd).

Any thoughts?  I don't have direct access to the MD3000i
unfortunately... this is from a CentOS 5.2 system btw.

Ray

Re: [dm-devel] multipathd: sdc: readsector0 checker reports path is down

[Konrad Rzeszutek]

On Thu, Jan 08, 2009 at 11:31:38AM -0800, Ray Van Dolson wrote:
> I'm still getting the hang of iSCSI and multipath, so bear with me if
> this is a FAQ that I've missed...
> 
> I have a host attaching to an MD3000i via iSCSI/dm-multipath that is
> working but showing a lot of the following errors in syslog:
> 
>   multipathd: sdc: readsector0 checker reports path is down

You are using the wrong path checker, wrong path priority program,
and no hardware path checker :-(

Here is what you need in your multipath.conf file:

 #   
    # DELL :D3000i :: Active-Passive RDAC
    # Note: The same as the IBM DS3300
    #   
    device {
        vendor          "DELL"
        product         "MD3000i"
        product_blacklist   "Universal Xport"
        features        "1 queue_if_no_path"
        path_grouping_policy    group_by_prio
        hardware_handler    "1 rdac"
        path_checker        rdac
        prio            "rdac"
        failback        immediate
    }   


Thought you might need to replace the 'prio	"rdac"' with
'prio_callout	"/sbin/mpath_prio_rdac" as I think RHEL5/CentOS5 uses
the old calling convention.

This should get rid of your down errors and show your passive/active path
properly.

Re: multipathd: sdc: readsector0 checker reports path is down

[Chandra Seetharaman]

On Thu, 2009-01-08 at 11:31 -0800, Ray Van Dolson wrote:
> I'm still getting the hang of iSCSI and multipath, so bear with me if
> this is a FAQ that I've missed...
> 
> I have a host attaching to an MD3000i via iSCSI/dm-multipath that is
> working but showing a lot of the following errors in syslog:
> 
>   multipathd: sdc: readsector0 checker reports path is down
you should be using the rdac path checker. In addition to that you
should be using rdac hardware handler and tpc priority checker.

If you have and entry for your MD3000i in /etc/multipath.conf, replace
it with this. If not add this
-------------------
devices {
        device {
                vendor                  "DELL"
                product                 "MD3000i"
                hardware_handler        "1 rdac"
                path_checker            rdac
                failback                immediate
                path_grouping_policy    group_by_prio
                no_path_retry           queue
                prio_callout            "/sbin/mpath_prio_tpc /dev/%n"
        }
}
---------------------

After this change you have to do:
- multipath -F
- service multipathd restart

Let me know how it goes.
> 
> My initial impression is that sdc is likely the "passive" path (for
> failover) whereas sdd is my active path:

yes.

> 
>   # multipath -ll
>   sdc: checker msg is "readsector0 checker reports path is down"
>   mpath2 (36001ec9000f338370000000000000000) dm-2 DELL,Universal Xport
>   [size=20M][features=0][hwhandler=0]
>   \_ round-robin 0 [prio=0][enabled]
>    \_ 4:0:0:31 sdf 8:80  [active][ready] 
>   mpath1 (36001ec9000f337dd0000000000000000) dm-1 DELL,Universal Xport
>   [size=20M][features=0][hwhandler=0]
>   \_ round-robin 0 [prio=0][enabled]
>    \_ 3:0:0:31 sde 8:64  [active][ready] 
>   mpath0 (36001ec9000f337dd0000065b491854a4) dm-0 DELL,MD3000i
>   [size=150G][features=0][hwhandler=1 rdac]
>   \_ round-robin 0 [prio=3][active]
>    \_ 3:0:0:0  sdd 8:48  [active][ready] 
>   \_ round-robin 0 [prio=0][enabled]
>    \_ 4:0:0:0  sdc 8:32  [failed][faulty]
> 
>   # iscsiadm -m discovery -t sendtargets -p 192.168.132.101:3260
>   0.0.0.0:3260,1 iqn.1984-05.com.dell:powervault.6001ec9000f337dd0000000048bdc862
>   0.0.0.0:3260,2 iqn.1984-05.com.dell:powervault.6001ec9000f337dd0000000048bdc862
>   192.168.132.101:3260,1 iqn.1984-05.com.dell:powervault.6001ec9000f337dd0000000048bdc862
>   192.168.132.102:3260,2 iqn.1984-05.com.dell:powervault.6001ec9000f337dd0000000048bdc862
> 
>   # iscsiadm -m session -P 1
>   Target: iqn.1984-05.com.dell:powervault.6001ec9000f337dd0000000048bdc862
>     Current Portal: 192.168.132.101:3260,1
>     Persistent Portal: 192.168.132.101:3260,1
>       **********
>       Interface:
>       **********
>       Iface Name: default
>       Iface Transport: tcp
>       Iface Initiatorname: iqn.1994-05.com.redhat:2738e0d3bd21
>       Iface IPaddress: 192.168.132.202
>       Iface HWaddress: default
>       Iface Netdev: default
>       SID: 1
>       iSCSI Connection State: LOGGED IN
>       iSCSI Session State: Unknown
>       Internal iscsid Session State: NO CHANGE
>     Current Portal: 192.168.132.102:3260,2
>     Persistent Portal: 192.168.132.102:3260,2
>       **********
>       Interface:
>       **********
>       Iface Name: default
>       Iface Transport: tcp
>       Iface Initiatorname: iqn.1994-05.com.redhat:2738e0d3bd21
>       Iface IPaddress: 192.168.132.202
>       Iface HWaddress: default
>       Iface Netdev: default
>       SID: 2
>       iSCSI Connection State: LOGGED IN
>       iSCSI Session State: Unknown
>       Internal iscsid Session State: NO CHANGE
> 
> sde and sdf are the silly 20MB partition that the MD3000 creates and
> I'm not using either.
> 
> My ultimate goal is to not have so many of the path is down errors
> showing up in my syslog file (hundreds and hundreds of them).  I'm not
> sure if their presence is normal (indicative of a passive path) or
> indicative of a problem.
> 
> Also, I'm not clear as to how I should identify which target
> corresponds with which drive (sdc or sdd).
> 
> Any thoughts?  I don't have direct access to the MD3000i
> unfortunately... this is from a CentOS 5.2 system btw.
> 
> Ray
> 
> --~--~---------~--~----~------------~-------~--~----~
> You received this message because you are subscribed to the Google Groups "open-iscsi" group.
> To post to this group, send email to open-iscsi@googlegroups.com
> To unsubscribe from this group, send email to open-iscsi+unsubscribe@googlegroups.com
> For more options, visit this group at http://groups.google.com/group/open-iscsi
> -~----------~----~----~----~------~----~------~--~---
> 

Re: multipathd: sdc: readsector0 checker reports path is down

[Ray Van Dolson]

On Thu, Jan 08, 2009 at 11:58:52AM -0800, Chandra Seetharaman wrote:
> 
> 
> On Thu, 2009-01-08 at 11:31 -0800, Ray Van Dolson wrote:
> > I'm still getting the hang of iSCSI and multipath, so bear with me if
> > this is a FAQ that I've missed...
> > 
> > I have a host attaching to an MD3000i via iSCSI/dm-multipath that is
> > working but showing a lot of the following errors in syslog:
> > 
> >   multipathd: sdc: readsector0 checker reports path is down
> you should be using the rdac path checker. In addition to that you
> should be using rdac hardware handler and tpc priority checker.
> 
> If you have and entry for your MD3000i in /etc/multipath.conf, replace
> it with this. If not add this
> -------------------
> devices {
>         device {
>                 vendor                  "DELL"
>                 product                 "MD3000i"
>                 hardware_handler        "1 rdac"
>                 path_checker            rdac
>                 failback                immediate
>                 path_grouping_policy    group_by_prio
>                 no_path_retry           queue
>                 prio_callout            "/sbin/mpath_prio_tpc /dev/%n"
>         }
> }
> ---------------------
> 
> After this change you have to do:
> - multipath -F
> - service multipathd restart
> 
> Let me know how it goes.

This appears to have done the trick.  So, I guess the default path
checker is readsector0 and rdac is another method -- I imagine it
directly talks to my MD3000i to determine path status?

> > 
> > My initial impression is that sdc is likely the "passive" path (for
> > failover) whereas sdd is my active path:
> 
> yes.

Thanks much!

Ray

Re: multipathd: sdc: readsector0 checker reports path is down

[Ray Van Dolson]

On Thu, Jan 08, 2009 at 11:56:43AM -0800, Konrad Rzeszutek wrote:
> 
> On Thu, Jan 08, 2009 at 11:31:38AM -0800, Ray Van Dolson wrote:
> > I'm still getting the hang of iSCSI and multipath, so bear with me if
> > this is a FAQ that I've missed...
> > 
> > I have a host attaching to an MD3000i via iSCSI/dm-multipath that is
> > working but showing a lot of the following errors in syslog:
> > 
> >   multipathd: sdc: readsector0 checker reports path is down
> 
> You are using the wrong path checker, wrong path priority program,
> and no hardware path checker :-(
> 
> Here is what you need in your multipath.conf file:
> 
>  #   
>     # DELL :D3000i :: Active-Passive RDAC
>     # Note: The same as the IBM DS3300
>     #   
>     device {
>         vendor          "DELL"
>         product         "MD3000i"
>         product_blacklist   "Universal Xport"
>         features        "1 queue_if_no_path"
>         path_grouping_policy    group_by_prio
>         hardware_handler    "1 rdac"
>         path_checker        rdac
>         prio            "rdac"
>         failback        immediate
>     }   
> 
> 
> Thought you might need to replace the 'prio	"rdac"' with
> 'prio_callout	"/sbin/mpath_prio_rdac" as I think RHEL5/CentOS5 uses
> the old calling convention.
> 
> This should get rid of your down errors and show your passive/active path
> properly.

Thanks.  I had to remove the features line -- got an Unrecognised
feature request error otherwise.

Thanks again,
Ray

Re: [dm-devel] Re: multipathd: sdc: readsector0 checker reports path is down

[Chandra Seetharaman]

On Thu, 2009-01-08 at 12:10 -0800, Ray Van Dolson wrote:
> On Thu, Jan 08, 2009 at 11:58:52AM -0800, Chandra Seetharaman wrote:
> > 
> > 
> > On Thu, 2009-01-08 at 11:31 -0800, Ray Van Dolson wrote:
> > > I'm still getting the hang of iSCSI and multipath, so bear with me if
> > > this is a FAQ that I've missed...
> > > 
> > > I have a host attaching to an MD3000i via iSCSI/dm-multipath that is
> > > working but showing a lot of the following errors in syslog:
> > > 
> > >   multipathd: sdc: readsector0 checker reports path is down
> > you should be using the rdac path checker. In addition to that you
> > should be using rdac hardware handler and tpc priority checker.
> > 
> > If you have and entry for your MD3000i in /etc/multipath.conf, replace
> > it with this. If not add this
> > -------------------
> > devices {
> >         device {
> >                 vendor                  "DELL"
> >                 product                 "MD3000i"
> >                 hardware_handler        "1 rdac"
> >                 path_checker            rdac
> >                 failback                immediate
> >                 path_grouping_policy    group_by_prio
> >                 no_path_retry           queue
> >                 prio_callout            "/sbin/mpath_prio_tpc /dev/%n"
> >         }
> > }
> > ---------------------
> > 
> > After this change you have to do:
> > - multipath -F
> > - service multipathd restart
> > 
> > Let me know how it goes.
> 
> This appears to have done the trick.  So, I guess the default path
> checker is readsector0 and rdac is another method -- I imagine it
> directly talks to my MD3000i to determine path status?

Yes, it sends a C9 inquiry to the storage controller and determines the
state.
> 
> > > 
> > > My initial impression is that sdc is likely the "passive" path (for
> > > failover) whereas sdd is my active path:
> > 
> > yes.
> 
> Thanks much!
> 
> Ray
> 
> --
> dm-devel mailing list
> dm-devel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org
> https://www.redhat.com/mailman/listinfo/dm-devel