B_Segurson's Posts

B_Segurson's Posts

I don't think you will get an official answer from Dell until the product is ready to release.  Michael Dell said something like this in the recent MD3000i announcement in his webcast.  Dell's pol... See more...
I don't think you will get an official answer from Dell until the product is ready to release.  Michael Dell said something like this in the recent MD3000i announcement in his webcast.  Dell's policy is to announce when they are ready to deliver.   That being said - you can answer your own question on "if" by noting that Dell is keeping up with disk technology on an ongoing basis.   The "when" question will be dependent on getting any new generation bugs resolved, and reliability up to a high standard before starting to ship.
The SCSI Management Module (SMM) does not support clustering.  SEMM does.  The difference is that the SEMM contains an expander chip that is essentially a bus repeater, but isolates the busses from t... See more...
The SCSI Management Module (SMM) does not support clustering.  SEMM does.  The difference is that the SEMM contains an expander chip that is essentially a bus repeater, but isolates the busses from the two servers, properly terminating them. 
The PowerVault 224F was designed as an expansion array behind a PowerVault 660F and works with FC-AL protocol.  However, you can try direct connecting to your host Fibre Channel HBA (QL2200, I recall)... See more...
The PowerVault 224F was designed as an expansion array behind a PowerVault 660F and works with FC-AL protocol.  However, you can try direct connecting to your host Fibre Channel HBA (QL2200, I recall).  You will not have hardware RAID capabilities and each of your disks will appear separately in your O/S.
Yellowpark,  The Dell part numbers you provided are for two cards, Part Number 3065E.   The last seven digits (983-A965 for one of the cards) give date of build and serial number identification.   ... See more...
Yellowpark,  The Dell part numbers you provided are for two cards, Part Number 3065E.   The last seven digits (983-A965 for one of the cards) give date of build and serial number identification.   For clustering you need a SEMM card, and the SEMM character set should appear somewhere on the card.   There is a jumper needed. It is labelled "Forced Join", and may be J8, if my memory is correct. Set this before turning on your system. The jumper is needed for both cards to share the same SCSI bus on the backplane.  If not there, the default is split bus mode and you will only see half the drives through each interface card.    
FYI setting to Cluster mode, then attaching to the same controller gives two paths to the same drives, and the software cannot resolve the conflict.  Both paths will think they "own" the drive and wi... See more...
FYI setting to Cluster mode, then attaching to the same controller gives two paths to the same drives, and the software cannot resolve the conflict.  Both paths will think they "own" the drive and will write conflicting data. When attaching both SEMMs to the same controller card, use Split Bus Mode.
The added drives will only  be operated at U80 speeds, as that is the maximum speed of the PERC2QC RAID card. You did not state which interface card is in the PV200S.  If you have SMM's then there ... See more...
The added drives will only  be operated at U80 speeds, as that is the maximum speed of the PERC2QC RAID card. You did not state which interface card is in the PV200S.  If you have SMM's then there needs to be a terminator card or physical terminator on the SMM in the other slot.   Also you could well be dealing with a cable problem.  Look at your Array Manager logs to see if there are any SCSI sense key=0B, code=47 logged errors that would indicate a cable or termination issue.
The PV210S uses remote power enable from the server it is attached to.  Thus you must be cabled to the server first, then have it powered on before your PV210S will power up, even though the power sup... See more...
The PV210S uses remote power enable from the server it is attached to.  Thus you must be cabled to the server first, then have it powered on before your PV210S will power up, even though the power supplies are turned on.  Try it, if you have not already.
Why not back up your data,  create a RAID 1 with the two larger drives, and restore to that drive?  RAID 1 defaults to the capacity of the smaller drive, so rather than maintaining an image of the... See more...
Why not back up your data,  create a RAID 1 with the two larger drives, and restore to that drive?  RAID 1 defaults to the capacity of the smaller drive, so rather than maintaining an image of the original 73GB drives (assuming you swapped in the drives one at a time and let the RAID 1 set rebuild), you could double your capacity.
The sticker should be on the outside sufrace of the connector bulkhead, near one of the SCSI connectors.   Only SEMM is used for clustering.
Saw a similar problem several years back.  Check to see what speed your drives are negotiating at (can be seen at boot time with most controllers).   If you see the speed going down as the ID gets hi... See more...
Saw a similar problem several years back.  Check to see what speed your drives are negotiating at (can be seen at boot time with most controllers).   If you see the speed going down as the ID gets higher then it could be a cable problem or lack of SCSI termination.  In my case it was the cable. Good luck in debugging this.
The PERC3/Di is the ROMB controller (Raid On MotherBoard) and should be connected to internal drives.  It is an Adaptec based controller.  The PERC3/DC and 3/DCL are external RAID controllers and are... See more...
The PERC3/Di is the ROMB controller (Raid On MotherBoard) and should be connected to internal drives.  It is an Adaptec based controller.  The PERC3/DC and 3/DCL are external RAID controllers and are properly used for your external PV200S storage.  The latter two are LSI based controllers, with a different RAID configuration than on the Adaptec controller. Migration is not simple if going from Adaptec to LSI. The key question with all these controllers in your system is where exactly have you plugged your PV220S cables in? To my knowledge, a new installation of Array Manager will pick up the RAID configurations previously established at the BIOS level.  I don't think you should have to destroy and rebuild RAID groups unless migrating from one brand of controller to another.
The PV200S had several options for termination.  If you had only one SCSI interface card, the other SCSI interface card slot should contain a terminator card. If you had SMM cards (see label n... See more...
The PV200S had several options for termination.  If you had only one SCSI interface card, the other SCSI interface card slot should contain a terminator card. If you had SMM cards (see label near SCSI connector) and only one connection to a PERC3/DC (which terminates its end of the bus), then a plug in terminator module is needed on the unused SMM card interface connector. If you had SEMM cards then the card terminates the SCSI bus before going through some bus expander logic.  The other SCSI interface card slot must also contain an SEMM card. if operating in split bus mode with SMM cards, the split bus module terminates the bus. There should be a manual online, if you no longer have one with the unit.
With all eight drives indicating predictive failures it is a fairly good bet that the cause is not the drives themselves.  What were the specific messages in Array Manager or Event Log?  Were there a... See more...
With all eight drives indicating predictive failures it is a fairly good bet that the cause is not the drives themselves.  What were the specific messages in Array Manager or Event Log?  Were there any disk related messages prior to the predictive failure message?   First your PERC card must be a /DC or /QC and the PV200S must be on a different channel than the one used for your internal drives.   Then you need to make sure your PV200S has the SCSI bus properly terminated, especially if you have SMM interface cards.  You also need to ensure you have a good SCSI cable.   There were problems a while back in the PERC software and in Array Manager software that could give a false predictive failure indication.  Make sure the firmware and software are at the latest levels available from the web.  I recall  the minimum levels to get past these problems were 1.98 for PERC3/DC and 3.4 for Array Manager.  I could be wrong on these, so check support.dell.com for latest downloads.
Connected to a PERC4/DC, yes it will work - at 80MB/s or 160MB/s depending on the drives and SEMM..  Do not however, put U320 capable SCSI drives in it (especially position 0 ), as there may be some s... See more...
Connected to a PERC4/DC, yes it will work - at 80MB/s or 160MB/s depending on the drives and SEMM..  Do not however, put U320 capable SCSI drives in it (especially position 0 ), as there may be some significant delay during power up while drives and controller negotiate a speed that might not be supported by the SEMM card.
You must plan on splitting your RAID group across the two sides of the PV221S, right?  There are 7 disks on a side when you split the backplane.  Otherwise you could use joined bus mode (a switch sett... See more...
You must plan on splitting your RAID group across the two sides of the PV221S, right?  There are 7 disks on a side when you split the backplane.  Otherwise you could use joined bus mode (a switch setting).   Did you order a PERC4/DC to go with it to connect to your server?
Your message is confusing.  Ther CERC SATA 6 channel will support the hard drives in a mirrored RAID 1 configuration.  You do not need another card, if I've interpreted your question correctly.
PERC4ei (or PERC4e/Di) is the RAID on motherboard (ROMB) version located on the PCI-Express bus on the riser card.  It is used to control internal drives.   The PERC4/DC has a PCI-X bus attach... See more...
PERC4ei (or PERC4e/Di) is the RAID on motherboard (ROMB) version located on the PCI-Express bus on the riser card.  It is used to control internal drives.   The PERC4/DC has a PCI-X bus attachment, and is an add-in card using a slot on the riser.  It can support either internal or external drives.  The /DC is the PERC4 card that supports clustered applications of the PV220S SCSI disk array.   If you are only using internal drives the PERC4ei may be the way to go.
lukanicbrian - A SCSI sense key and code of FF is invalid and represents some sort of transient condition that is not being reported correctly.  If the disk is offline, then I would suggest you bring... See more...
lukanicbrian - A SCSI sense key and code of FF is invalid and represents some sort of transient condition that is not being reported correctly.  If the disk is offline, then I would suggest you bring it back online and let it rebuild or go through a consistency check.  There is not enough information to diagnose the cause of the problem.  It may not be hardware at all.
The battery backed cache retains data through power glitches and outages of limited duration.   The original question was if the cache memory remains valid after a controller failure while replac... See more...
The battery backed cache retains data through power glitches and outages of limited duration.   The original question was if the cache memory remains valid after a controller failure while replacing the motherboard.    There is no way that memory contents' validity can be guaranteed while replacing components.   The battery backed cache is a DIMM plug in module with battery onboard.  As with any card or DIMM it is possible to inadvertently short out pins when removing and reinserting the DIMM.  Therefore there is absolutely no guarantee of data validity through component replacement operations.     Battery backed cache is an operational feature, for times when the system is functional and no component replacement is in progress.   As mentioned it preserves unwrittern data through power failures.
Sounds fine.  Go for it.