Openserver 5.0.7 and hot-plug tape - SCO

This is a discussion on Openserver 5.0.7 and hot-plug tape - SCO ; Hi all, I'm having some problems with an hot-plug tape drive. The server is an HP Proliant DL380 G4 and the tape (HP DAT 72) is connected to the embedded Smart Array 6i controller. Operating system is SCO OpenServer 5 ...

+ Reply to Thread
Results 1 to 6 of 6

Thread: Openserver 5.0.7 and hot-plug tape

  1. Openserver 5.0.7 and hot-plug tape

    Hi all,

    I'm having some problems with an hot-plug tape drive.

    The server is an HP Proliant DL380 G4 and the tape (HP DAT 72) is
    connected to the embedded Smart Array 6i controller.
    Operating system is SCO OpenServer 5 (Release = 3.2v5.0.7, KernelID =
    2003-02-18) with EFS 5.64a loaded.

    The backup command is a simple

    cat /tmp/list_of_files | cpio -ocC32768 -O /dev/rStp0

    and the error messages reported are:

    -- from syslog --
    Jul 26 04:19:09 scounix WARNING: ciss : SCSI bus speed
    downshifted, SCSI port 2
    Jul 26 04:19:09 scounix NOTICE: ciss: command completed with error
    status = 0x0000000B for controller at bus =4 ,slot =0, func =0 sense info =
    -- end --

    About 3 weeks ago we called HP's tech support.
    They changed the tape drive and everything was going fine, but now the
    problem arised again with the same behaviour. So I guess the problem is
    elsewhere (or they changed the tape with a defective one ).

    Anyone with similar problem around here?
    Where can I find an explanation of "error status 0x0000000B"?

    Thanks in advance for any help...
    Davide

  2. Re: Openserver 5.0.7 and hot-plug tape


    DavideC wrote:
    > Hi all,
    >
    > I'm having some problems with an hot-plug tape drive.
    >
    > The server is an HP Proliant DL380 G4 and the tape (HP DAT 72) is
    > connected to the embedded Smart Array 6i controller.
    > Operating system is SCO OpenServer 5 (Release = 3.2v5.0.7, KernelID =
    > 2003-02-18) with EFS 5.64a loaded.
    >
    >


    Don't connect the tape drive to a CISS controller. Previous experience
    has shown me that they don't like tape drives of any kind, they much
    prefer external storage array units. I had to use a seperate Adaptec
    card for my DLT unit. I think they have been previous posts regarding
    DLTs and the CISS controller.


  3. Re: Openserver 5.0.7 and hot-plug tape


    ----- Original Message -----
    From:
    Newsgroups: comp.unix.sco.misc
    To:
    Sent: Wednesday, August 03, 2005 11:37 AM
    Subject: Re: Openserver 5.0.7 and hot-plug tape


    >
    > DavideC wrote:
    >> Hi all,
    >>
    >> I'm having some problems with an hot-plug tape drive.
    >>
    >> The server is an HP Proliant DL380 G4 and the tape (HP DAT 72) is
    >> connected to the embedded Smart Array 6i controller.
    >> Operating system is SCO OpenServer 5 (Release = 3.2v5.0.7, KernelID =
    >> 2003-02-18) with EFS 5.64a loaded.
    >>
    >>

    >
    > Don't connect the tape drive to a CISS controller. Previous experience
    > has shown me that they don't like tape drives of any kind, they much
    > prefer external storage array units. I had to use a seperate Adaptec
    > card for my DLT unit. I think they have been previous posts regarding
    > DLTs and the CISS controller.


    I found this out myself the hard way with a recent DL380 G4 / sco 5.0.7
    install.

    I sent them an adaptec card.

    Brian K. White -- brian@aljex.com -- http://www.aljex.com/bkw/
    +++++[>+++[>+++++>+++++++<<-]<-]>>+.>.+++++.+++++++.-.[>+<---]>++.
    filePro BBx Linux SCO FreeBSD #callahans Satriani Filk!


  4. Re: Openserver 5.0.7 and hot-plug tape

    DavideC wrote:
    > Hi all,
    >
    > I'm having some problems with an hot-plug tape drive.
    >
    > The server is an HP Proliant DL380 G4 and the tape (HP DAT 72) is
    > connected to the embedded Smart Array 6i controller.
    > Operating system is SCO OpenServer 5 (Release = 3.2v5.0.7, KernelID =
    > 2003-02-18) with EFS 5.64a loaded.
    >
    > The backup command is a simple
    >
    > cat /tmp/list_of_files | cpio -ocC32768 -O /dev/rStp0
    >
    > and the error messages reported are:
    >
    > -- from syslog --
    > Jul 26 04:19:09 scounix WARNING: ciss : SCSI bus speed
    > downshifted, SCSI port 2
    > Jul 26 04:19:09 scounix NOTICE: ciss: command completed with error
    > status = 0x0000000B for controller at bus =4 ,slot =0, func =0 sense info =
    > -- end --
    >
    > About 3 weeks ago we called HP's tech support.
    > They changed the tape drive and everything was going fine, but now the
    > problem arised again with the same behaviour. So I guess the problem is
    > elsewhere (or they changed the tape with a defective one ).
    >
    > Anyone with similar problem around here?
    > Where can I find an explanation of "error status 0x0000000B"?
    >
    > Thanks in advance for any help...
    > Davide


    Hi Davide,

    HP's EFS instructions say you can use the Hot Swap tape and that you
    need to use Bus=4, Lun=0, scsi id=0.

    The following is not exactly your problem but I have a client with a
    DL380/G4 with a hot swap configured as above. It is running 5.0.7 with
    all relevant updates (MP3 etc) and EFS5.64a. The only problem we have
    had is that we have had some occurrences of the driver not responding.
    You can issue a tape -s status command and you get back a message
    stating that their is no tape in the tape drive ie: no-medium.

    When it first occurred HP swapped the tape drive out believing it was a
    faulty drive. The drive had been in for around 4 months and had not
    missed a beat. The same problem occurred again a few weeks after it had
    been replaced. At this point I had the user pull the drive out and put
    it back in. Being a hot swap device I wanted to see if the unit would
    reset. It did. This proved that it was not a drive (hardware), issue as
    such. You may be able to do this as a work around and prove if you are
    having a hardware of software problem.

    So now if the drive plays up the end user just pulls out the drive and
    re-inserts it.

    We have been trying to get a resolution from HP but they say we should
    have the latest EFS, BIOS updates etc installed.
    We have upgraded the bios and firmware of all components and have tried
    installing EFS 5.66a and 5.68a. However we do not know if either of
    these later EFS's would fix the problem since when we install them we get

    %adapter 0x4000-0x4100 7 - type=ciss ha=0 PCI slot=0 (rev 5.68a)
    %casm 0x1800-0x18FF 7 - hp ProLiant Advanced Server Management
    Driver (rev 6.68a)
    WARNING: bcme : This driver does not support ASF. Instructions on how
    to disable this feature are in the readme.txt file included in the
    BRCMbcme package.
    %bcme0 0xFFFF-0xFFFE - - Get adapter info failed
    %bcme0 0x0000-0xFFFF - - Adapter not present

    Basically no NIC.

    Have you tried any later EFS's and or upgrading the firmware etc?
    If you have did you get a message re AFS?

    No one at HP, as yet, has been able to enlighten us as to how to rectify
    the AFS error and no one can tell me what ASF means.

    I also have clients with external DAT72 drives connected to an ad-on
    scsi card. These servers and tape drivers have never had a problem.

    I know this is not exactly like your problem and it is not an answer to
    it but it may help in resolving it

    Also as far as I know you cannot connect the hot swap tape to a add on
    scsi card. At least that is what HP have told me. To use an add on card
    they would have to use an external tape drive.

    If I get a resolution to my problem I will post info for you here.

    Good luck!

  5. Re: Openserver 5.0.7 and hot-plug tape


    ----- Original Message -----
    From: "Craig Green"
    Newsgroups: comp.unix.sco.misc
    To:
    Sent: Tuesday, August 09, 2005 12:52 AM
    Subject: Re: Openserver 5.0.7 and hot-plug tape


    > DavideC wrote:
    >> Hi all,
    >>
    >> I'm having some problems with an hot-plug tape drive.
    >>
    >> The server is an HP Proliant DL380 G4 and the tape (HP DAT 72) is
    >> connected to the embedded Smart Array 6i controller.
    >> Operating system is SCO OpenServer 5 (Release = 3.2v5.0.7, KernelID =
    >> 2003-02-18) with EFS 5.64a loaded.
    >>
    >> The backup command is a simple
    >>
    >> cat /tmp/list_of_files | cpio -ocC32768 -O /dev/rStp0
    >>
    >> and the error messages reported are:
    >>
    >> -- from syslog --
    >> Jul 26 04:19:09 scounix WARNING: ciss : SCSI bus speed
    >> downshifted, SCSI port 2
    >> Jul 26 04:19:09 scounix NOTICE: ciss: command completed with error
    >> status = 0x0000000B for controller at bus =4 ,slot =0, func =0 sense info
    >> =
    >> -- end --
    >>
    >> About 3 weeks ago we called HP's tech support.
    >> They changed the tape drive and everything was going fine, but now the
    >> problem arised again with the same behaviour. So I guess the problem is
    >> elsewhere (or they changed the tape with a defective one ).
    >>
    >> Anyone with similar problem around here?
    >> Where can I find an explanation of "error status 0x0000000B"?
    >>
    >> Thanks in advance for any help...
    >> Davide

    >
    > Hi Davide,
    >
    > HP's EFS instructions say you can use the Hot Swap tape and that you need
    > to use Bus=4, Lun=0, scsi id=0.
    >
    > The following is not exactly your problem but I have a client with a
    > DL380/G4 with a hot swap configured as above. It is running 5.0.7 with all
    > relevant updates (MP3 etc) and EFS5.64a. The only problem we have had is
    > that we have had some occurrences of the driver not responding. You can
    > issue a tape -s status command and you get back a message stating that
    > their is no tape in the tape drive ie: no-medium.
    >
    > When it first occurred HP swapped the tape drive out believing it was a
    > faulty drive. The drive had been in for around 4 months and had not
    > missed a beat. The same problem occurred again a few weeks after it had
    > been replaced. At this point I had the user pull the drive out and put it
    > back in. Being a hot swap device I wanted to see if the unit would reset.
    > It did. This proved that it was not a drive (hardware), issue as such. You
    > may be able to do this as a work around and prove if you are having a
    > hardware of software problem.
    >
    > So now if the drive plays up the end user just pulls out the drive and
    > re-inserts it.
    >
    > We have been trying to get a resolution from HP but they say we should
    > have the latest EFS, BIOS updates etc installed.
    > We have upgraded the bios and firmware of all components and have tried
    > installing EFS 5.66a and 5.68a. However we do not know if either of these
    > later EFS's would fix the problem since when we install them we get
    >
    > %adapter 0x4000-0x4100 7 - type=ciss ha=0 PCI slot=0 (rev 5.68a)
    > %casm 0x1800-0x18FF 7 - hp ProLiant Advanced Server Management
    > Driver (rev 6.68a)
    > WARNING: bcme : This driver does not support ASF. Instructions on how to
    > disable this feature are in the readme.txt file included in the BRCMbcme
    > package.
    > %bcme0 0xFFFF-0xFFFE - - Get adapter info failed
    > %bcme0 0x0000-0xFFFF - - Adapter not present
    >
    > Basically no NIC.
    >
    > Have you tried any later EFS's and or upgrading the firmware etc?
    > If you have did you get a message re AFS?
    >
    > No one at HP, as yet, has been able to enlighten us as to how to rectify
    > the AFS error and no one can tell me what ASF means.


    I installed EFS 5.66a on a DL380 G4 as part of my initial system load.
    I vaguely barely, not really I guess, remember seeing some message like
    that. I don't remember if I actually had to do anything about it though. The
    bcme nic does work and has been in production for a few months now.

    .....
    ok you don't have any excuse for asking this
    "bcme asf" punched into google brings up this as the very first link:
    http://wdb1.sco.com/kb/showta?taid=126083

    which tells you all about it.

    In my case, I dont remember doing any of that but I might have. It was a
    hectic week dealing with lots of compaq weirdness. Give me white boxes over
    this proprietary stuff any day.
    The 5th floppy disk created by EFS 5.66a might have had it's own notes on
    the subject.
    Before I googled this up just now, I would have said either I didn't do
    anything, or that if I did anything it was in the bios or possibly via the
    special boot cd that came with the server, not a dos floppy.
    In fact, the more I think about it, the more sure I am I never made a dos
    floppy nor entered any commands like that. There must be another way to turn
    off asf, some way that's fairly intuitive since apparently I found it and
    did it with little enough trouble that I didn't bother remembering the
    details.
    I'm looking at the box now and I definitely have 5.66a installed, and
    definitely using the bcme driver from the efs not the version that comes
    stock or can be downloaded from sco.

    [...]
    Aug 9 00:55:44 advunix %casm 0x1800-0x18FF 7 - hp ProLiant Advanced
    Ser
    ver Management Driver (rev 5.66a)
    Aug 9 00:55:44 advunix %bcme0 - 5 - chip=BCM5704
    mem=FDDF000
    0 phy=BCM5704 addr=00:0e:7f:20:af:61
    [...]
    Aug 9 00:55:44 advunix NOTICE: bcme0 (slot:1 port:1): NIC Link is Up
    100Mbps fu
    ll duplex, receive & transmit flow control ON

    And looking elsewhere around the net it looks like the 5704 does have the
    asf feature so the topic did apply to this box.
    One thing I know I did manually in the bios that had to do with the nic was
    I disabled iLO.

    Brian K. White -- brian@aljex.com -- http://www.aljex.com/bkw/
    +++++[>+++[>+++++>+++++++<<-]<-]>>+.>.+++++.+++++++.-.[>+<---]>++.
    filePro BBx Linux SCO FreeBSD #callahans Satriani Filk!


  6. Re: Openserver 5.0.7 and hot-plug tape

    Brian K. White wrote:
    > ----- Original Message -----
    > From: "Craig Green"
    > Newsgroups: comp.unix.sco.misc
    > To:
    > Sent: Tuesday, August 09, 2005 12:52 AM
    > Subject: Re: Openserver 5.0.7 and hot-plug tape
    >
    >
    >
    >>DavideC wrote:
    >>
    >>>Hi all,
    >>>
    >>>I'm having some problems with an hot-plug tape drive.
    >>>
    >>>The server is an HP Proliant DL380 G4 and the tape (HP DAT 72) is
    >>>connected to the embedded Smart Array 6i controller.
    >>>Operating system is SCO OpenServer 5 (Release = 3.2v5.0.7, KernelID =
    >>>2003-02-18) with EFS 5.64a loaded.
    >>>
    >>>The backup command is a simple
    >>>
    >>>cat /tmp/list_of_files | cpio -ocC32768 -O /dev/rStp0
    >>>
    >>>and the error messages reported are:
    >>>
    >>>-- from syslog --
    >>>Jul 26 04:19:09 scounix WARNING: ciss : SCSI bus speed
    >>>downshifted, SCSI port 2
    >>>Jul 26 04:19:09 scounix NOTICE: ciss: command completed with error
    >>>status = 0x0000000B for controller at bus =4 ,slot =0, func =0 sense info
    >>>=
    >>>-- end --
    >>>
    >>>About 3 weeks ago we called HP's tech support.
    >>>They changed the tape drive and everything was going fine, but now the
    >>>problem arised again with the same behaviour. So I guess the problem is
    >>>elsewhere (or they changed the tape with a defective one ).
    >>>
    >>>Anyone with similar problem around here?
    >>>Where can I find an explanation of "error status 0x0000000B"?
    >>>
    >>>Thanks in advance for any help...
    >>>Davide

    >>
    >>Hi Davide,
    >>
    >>HP's EFS instructions say you can use the Hot Swap tape and that you need
    >>to use Bus=4, Lun=0, scsi id=0.
    >>
    >>The following is not exactly your problem but I have a client with a
    >>DL380/G4 with a hot swap configured as above. It is running 5.0.7 with all
    >>relevant updates (MP3 etc) and EFS5.64a. The only problem we have had is
    >>that we have had some occurrences of the driver not responding. You can
    >>issue a tape -s status command and you get back a message stating that
    >>their is no tape in the tape drive ie: no-medium.
    >>
    >>When it first occurred HP swapped the tape drive out believing it was a
    >>faulty drive. The drive had been in for around 4 months and had not
    >>missed a beat. The same problem occurred again a few weeks after it had
    >>been replaced. At this point I had the user pull the drive out and put it
    >>back in. Being a hot swap device I wanted to see if the unit would reset.
    >>It did. This proved that it was not a drive (hardware), issue as such. You
    >>may be able to do this as a work around and prove if you are having a
    >>hardware of software problem.
    >>
    >>So now if the drive plays up the end user just pulls out the drive and
    >>re-inserts it.
    >>
    >>We have been trying to get a resolution from HP but they say we should
    >>have the latest EFS, BIOS updates etc installed.
    >>We have upgraded the bios and firmware of all components and have tried
    >>installing EFS 5.66a and 5.68a. However we do not know if either of these
    >>later EFS's would fix the problem since when we install them we get
    >>
    >>%adapter 0x4000-0x4100 7 - type=ciss ha=0 PCI slot=0 (rev 5.68a)
    >>%casm 0x1800-0x18FF 7 - hp ProLiant Advanced Server Management
    >>Driver (rev 6.68a)
    >>WARNING: bcme : This driver does not support ASF. Instructions on how to
    >>disable this feature are in the readme.txt file included in the BRCMbcme
    >>package.
    >>%bcme0 0xFFFF-0xFFFE - - Get adapter info failed
    >>%bcme0 0x0000-0xFFFF - - Adapter not present
    >>
    >>Basically no NIC.
    >>
    >>Have you tried any later EFS's and or upgrading the firmware etc?
    >>If you have did you get a message re AFS?
    >>
    >>No one at HP, as yet, has been able to enlighten us as to how to rectify
    >>the AFS error and no one can tell me what ASF means.

    >
    >
    > I installed EFS 5.66a on a DL380 G4 as part of my initial system load.
    > I vaguely barely, not really I guess, remember seeing some message like
    > that. I don't remember if I actually had to do anything about it though. The
    > bcme nic does work and has been in production for a few months now.
    >
    > ....
    > ok you don't have any excuse for asking this
    > "bcme asf" punched into google brings up this as the very first link:
    > http://wdb1.sco.com/kb/showta?taid=126083
    >
    > which tells you all about it.
    >
    > In my case, I dont remember doing any of that but I might have. It was a
    > hectic week dealing with lots of compaq weirdness. Give me white boxes over
    > this proprietary stuff any day.
    > The 5th floppy disk created by EFS 5.66a might have had it's own notes on
    > the subject.
    > Before I googled this up just now, I would have said either I didn't do
    > anything, or that if I did anything it was in the bios or possibly via the
    > special boot cd that came with the server, not a dos floppy.
    > In fact, the more I think about it, the more sure I am I never made a dos
    > floppy nor entered any commands like that. There must be another way to turn
    > off asf, some way that's fairly intuitive since apparently I found it and
    > did it with little enough trouble that I didn't bother remembering the
    > details.
    > I'm looking at the box now and I definitely have 5.66a installed, and
    > definitely using the bcme driver from the efs not the version that comes
    > stock or can be downloaded from sco.
    >
    > [...]
    > Aug 9 00:55:44 advunix %casm 0x1800-0x18FF 7 - hp ProLiant Advanced
    > Ser
    > ver Management Driver (rev 5.66a)
    > Aug 9 00:55:44 advunix %bcme0 - 5 - chip=BCM5704
    > mem=FDDF000
    > 0 phy=BCM5704 addr=00:0e:7f:20:af:61
    > [...]
    > Aug 9 00:55:44 advunix NOTICE: bcme0 (slot:1 port:1): NIC Link is Up
    > 100Mbps fu
    > ll duplex, receive & transmit flow control ON
    >
    > And looking elsewhere around the net it looks like the 5704 does have the
    > asf feature so the topic did apply to this box.
    > One thing I know I did manually in the bios that had to do with the nic was
    > I disabled iLO.
    >
    > Brian K. White -- brian@aljex.com -- http://www.aljex.com/bkw/
    > +++++[>+++[>+++++>+++++++<<-]<-]>>+.>.+++++.+++++++.-.[>+<---]>++.
    > filePro BBx Linux SCO FreeBSD #callahans Satriani Filk!
    >

    Thanks for the info. Hopefully this will help.

    Yes, if you do a google now you do indeed got back some links. But I can
    guarantee you that this was not the case when I experienced the problem
    back on July 1. I see that the original post date for the sco TA is July
    1. But this would have been after I had the problem since I am in the
    land of Oz (Australia), around 14 hours ahead of the good old US. If
    only I had been doing this on or after the 2nd of July.

    Since then we have been talking solely with HP and relying upon them for
    a resolution. Yes, I know, silly me. I probably should have performed
    more searches since. What fascinates me is that HP have still not
    supplied an answer.

    The readme.txt file that the warning refers to is not on any of the EFS
    floppys or on the sco box.

    Turing off the iLO is not an option for us. So I guess I am going to
    have to try the info in the TA after I a least point HP in the direction
    of the SCO info. I really want a solution from HP, namely a driver that
    works, since once you start using non HP supplied drivers etc they
    basically say it is not there problem if something is not working correctly.

+ Reply to Thread