Adaptec 2940UW with Fujitsu MAN3184M
Locked
Adaptec 2940UW with Fujitsu MAN3184M
 
|
Hello,
I'm using Adaptec 2940UW with Fujitsu MAN3184M, round Ultra-320 terminated? cable (has a black box at the end), and a 68-pin to SCA-1 adapter. When the driver is loaded, it says "FAST-10 WIDE SCSI 20.0 MB/s ST (100 ns, offset 8)". 1. Is it possible to make this combination work at 40 MB/s? Should I recheck termination? 2. Can I use the drive at 160 MB/s if I buy a 29160 controller (with the adapter)? Thanks in advance, -- Baurzhan Ismagulov http://www.kz-easy.com/ _______________________________________________ [hidden email] mailing list http://lists.freebsd.org/mailman/listinfo/aic7xxx To unsubscribe, send any mail to "[hidden email]" |
Re: Adaptec 2940UW with Fujitsu MAN3184M
|
Hello,
The Adaptec 2940UW is Single Ended. With Single Ended, there are a number of things to consider when trying to go Ultra. It is possible that even though the 2940UW initially negotiates for Ultra, the domain validation forces it to the slower rate due to issues seen with the communications. The other thing to check is if the 2940UW BIOS has been forced to the slower rate. Also with Single Ended, the SCSI Cable Length can be critical. Depending on a number of factors, going more than 2 meters may add to the problem. The SCA-1 Adapter may also be adding to the problem unless it is known to be able to work at the Ultra burst rate in Single Ended. Also verify how you have the terminations. If these are the only two devices, the 2940UW must have its terminations turned on (automatic may be ok) and there should be a termination at the drive. I believe the Fujitsu MAN3184M is Ultra160 Low Voltage Differential (LVD), and capable of automatically going down to Single Ended communications from LVD communications, like almost all LVD devices, but its not being used to its full potential by doing so. Changing to the 29160 controller which is Low Voltage Dfferential (LVD) will help immensely in terms of being able to operate at a faster burst rate and being able to reliably support longer SCSI cable lengths. Regards, Marc D. Brooks Baurzhan Ismagulov wrote: >Hello, > >I'm using Adaptec 2940UW with Fujitsu MAN3184M, round Ultra-320 >terminated? cable (has a black box at the end), and a 68-pin to SCA-1 >adapter. > >When the driver is loaded, it says "FAST-10 WIDE SCSI 20.0 MB/s ST (100 >ns, offset 8)". > >1. Is it possible to make this combination work at 40 MB/s? Should I > recheck termination? > >2. Can I use the drive at 160 MB/s if I buy a 29160 controller (with the > adapter)? > >Thanks in advance, > > [hidden email] mailing list http://lists.freebsd.org/mailman/listinfo/aic7xxx To unsubscribe, send any mail to "[hidden email]" |
Re: Adaptec 2940UW with Fujitsu MAN3184M
|
|
Hello Marc,
Thanks for the prompt answer! Let's see what we can find out. On Wed, Feb 28, 2007 at 02:31:15PM -0800, Marc D. Brooks wrote: > The Adaptec 2940UW is Single Ended. With Single Ended, there are a > number of things to consider when trying to go Ultra. It is possible > that even though the 2940UW initially negotiates for Ultra, the domain > validation forces it to the slower rate due to issues seen with the > communications. Any way to see those at the Linux administrator level? > The other thing to check is if the 2940UW BIOS has been forced to the > slower rate. The card is branded Siemens-Nixdorf, I don't see the usual Ctrl-A prompt. I've seen older SNI controllers, all settings were accessible from the main BIOS. So, I guess I can't check this. That said, when the driver is loaded, it states the USER setting of 40 MB/s. Justin said that this is determined by EEPROM and /proc. I haven't set anything in /proc, so I guess we can rule out the BIOS setting limitation. The "when" part above is strange. Now I don't have the drive connected, aic7xxx is loaded, but I see neither kernel messages in log, nor /proc/aic7xxx. I'm running Debian sid's 2.4.18-4-k7. rmmod, modprobe don't help. At the last boot the module was also loaded, no messages, but rmmod, modprobe did help. The red LED named "DS1" is constantly lit. Should I throw the card away? Or could it be anything else (power supply???)? > Also with Single Ended, the SCSI Cable Length can be critical. > Depending on a number of factors, going more than 2 meters may add to > the problem. 1.2 m, four 68-pin connectors for drives, black box at the end saying "Foxconn" and "Ultra 320" (terminator?), the drive connected to the last entry before the terminator ("far" from the controller). > The SCA-1 Adapter may also be adding to the problem unless it is known > to be able to work at the Ultra burst rate in Single Ended. Hmm, specifically SE? The adapter claims to be "Ultra4 320/m SCSI compliant". I'm not sure what "/m" means. MHz? It doesn't have any active components. One resistor goes from SCA-2 to the jumper named "LED". There are also SYN, DLY, MTR, ID3-0 jumpers, none set. > Also verify how you have the terminations. If these are the only two > devices, the 2940UW must have its terminations turned on (automatic > may be ok) and there should be a termination at the drive. The disk is the only device on the bus. If the black box is a terminator, then I guess I don't need termination on the drive. The drive is branded Sun and doesn't have any jumpers. I don't know whether one of the jumpers on the adapter is for termination; I assume none. At the other end, the controller has three connectors, two internal and one external. I guess it should be terminated. There are six jumper-like pins, four named J3, and two named J4. http://download.adaptec.com/pdfs/user_guides/p2940uw_ug.pdf doesn't mention anything about them, but says the termination can be set in BIOS to Auto, Enabled, or Disabled. Is there a way to check the setting? > I believe the Fujitsu MAN3184M is Ultra160 Low Voltage Differential > (LVD), and capable of automatically going down to Single Ended > communications from LVD communications, like almost all LVD devices, but > its not being used to its full potential by doing so. Yes. > Changing to the 29160 controller which is Low Voltage Dfferential > (LVD) will help immensely in terms of being able to operate at a > faster burst rate and being able to reliably support longer SCSI cable > lengths. Do you think this will work properly with the adapter? I had stability problems with a similar adapter, AHA-2940UW, and the same drive (although that one was not claiming Ultra 320 compliancy). With kind regards, -- Baurzhan Ismagulov http://www.kz-easy.com/ _______________________________________________ [hidden email] mailing list http://lists.freebsd.org/mailman/listinfo/aic7xxx To unsubscribe, send any mail to "[hidden email]" |
Re: Adaptec 2940UW with Fujitsu MAN3184M
|
Baurzhan Ismagulov wrote:
> Hello Marc, > > Thanks for the prompt answer! Let's see what we can find out. > > On Wed, Feb 28, 2007 at 02:31:15PM -0800, Marc D. Brooks wrote: >> The Adaptec 2940UW is Single Ended. With Single Ended, there are a >> number of things to consider when trying to go Ultra. It is possible >> that even though the 2940UW initially negotiates for Ultra, the domain >> validation forces it to the slower rate due to issues seen with the >> communications. > > Any way to see those at the Linux administrator level? > at least with the "AIC-7899A U160/m" I am running you can see them by looking at /proc/scsi/aic7xxx/0, i.e. less /proc/scsi/aic7xxx/0 or cat /proc/scsi/aic7xxx/0 I believe the "User: " line is what the card bios is set to. I have no idea how/if the "Goal:" line can be modified. I believe the "Curr:" is what is currently being used to communicate with the device. I set the card bios on my system to 66MB/s and the info below is what proc gives. Target 0 Negotiation Settings User: 66.006MB/s transfers (33.003MHz DT, offset 127, 16bit) Goal: 66.006MB/s transfers (33.003MHz DT, offset 62, 16bit) Curr: 66.006MB/s transfers (33.003MHz DT, offset 62, 16bit) > >> The other thing to check is if the 2940UW BIOS has been forced to the >> slower rate. > > The card is branded Siemens-Nixdorf, I don't see the usual Ctrl-A > prompt. I've seen older SNI controllers, all settings were accessible > from the main BIOS. So, I guess I can't check this. > > That said, when the driver is loaded, it states the USER setting of 40 > MB/s. Justin said that this is determined by EEPROM and /proc. I haven't > set anything in /proc, so I guess we can rule out the BIOS setting > limitation. > did he say what in proc could be used to set it? or was he just indicating that is where to look to find what it is set to? > The "when" part above is strange. Now I don't have the drive connected, > aic7xxx is loaded, but I see neither kernel messages in log, nor > /proc/aic7xxx. should be /proc/scsi/aic7xxx/ I think. > I'm running Debian sid's 2.4.18-4-k7. rmmod, modprobe > don't help. At the last boot the module was also loaded, no messages, > but rmmod, modprobe did help. The red LED named "DS1" is constantly lit. > Should I throw the card away? Or could it be anything else (power > supply???)? > > -- Todd Denniston Crane Division, Naval Surface Warfare Center (NSWC Crane) Harnessing the Power of Technology for the Warfighter _______________________________________________ [hidden email] mailing list http://lists.freebsd.org/mailman/listinfo/aic7xxx To unsubscribe, send any mail to "[hidden email]" |
Re: Adaptec 2940UW with Fujitsu MAN3184M
|
|
Hello Todd,
On Thu, Mar 01, 2007 at 09:59:51AM -0600, Todd Denniston wrote: > >>The Adaptec 2940UW is Single Ended. With Single Ended, there are a > >>number of things to consider when trying to go Ultra. It is possible > >>that even though the 2940UW initially negotiates for Ultra, the domain > >>validation forces it to the slower rate due to issues seen with the > >>communications. > > > >Any way to see those at the Linux administrator level? > > at least with the "AIC-7899A U160/m" I am running you can see them by > looking at /proc/scsi/aic7xxx/0, i.e. I see the following: Adaptec AIC7xxx driver version: 7.0 Adaptec aic7880 Ultra SCSI adapter aic7880: Ultra Wide Channel A, SCSI Id=7, 16/253 SCBs Allocated SCBs: 4, SG List Length: 128 Serial EEPROM: 0x0338 0x0338 0x0338 0x0338 0x0338 0x0338 0x0338 0x0238 0x0338 0x0338 0x0338 0x0338 0x0338 0x0338 0x0338 0x0338 0x10b6 0x00db 0x2807 0x0010 0xff00 0xffff 0xffff 0xffff 0xffff 0xffff 0xffff 0xffff 0xffff 0xffff 0x00ff 0x6c1e Target 0 Negotiation Settings User: 40.000MB/s transfers (20.000MHz, offset 127, 16bit) Goal: 20.000MB/s transfers (10.000MHz, offset 8, 16bit) Curr: 20.000MB/s transfers (10.000MHz, offset 8, 16bit) Channel A Target 0 Lun 0 Settings Commands Queued 58 Commands Active 0 Command Openings 8 Max Tagged Openings 8 Device Queue Frozen Count 0 ... I see that the bus frequency is limited to 10 MHz, but I don't see why (which "issues seen with the communications" were the reason for that) and what I could do to change that. > did he say what in proc could be used to set it? or was he just indicating > that is where to look to find what it is set to? No. Yes. But I have 20 MHz anyway, that's the maximum the card can do and that I want to reach. > >The "when" part above is strange. Now I don't have the drive connected, > >aic7xxx is loaded, but I see neither kernel messages in log, nor > >/proc/aic7xxx. > > should be /proc/scsi/aic7xxx/ I think. Ah, thanks! Any further ideas? With kind regards, -- Baurzhan Ismagulov http://www.kz-easy.com/ _______________________________________________ [hidden email] mailing list http://lists.freebsd.org/mailman/listinfo/aic7xxx To unsubscribe, send any mail to "[hidden email]" |
Re: Adaptec 2940UW with Fujitsu MAN3184M
|
|
In reply to this post by Baurzhan Ismagulov
Hello Baurzhan,
I'll try to answer each of the sections below. Baurzhan Ismagulov wrote: >Hello Marc, > >Thanks for the prompt answer! Let's see what we can find out. > >On Wed, Feb 28, 2007 at 02:31:15PM -0800, Marc D. Brooks wrote: > > >>The Adaptec 2940UW is Single Ended. With Single Ended, there are a >>number of things to consider when trying to go Ultra. It is possible >>that even though the 2940UW initially negotiates for Ultra, the domain >>validation forces it to the slower rate due to issues seen with the >>communications. >> >> > >Any way to see those at the Linux administrator level? > > > > specifications for the device and the HBA. Some BIOS implementations may report the Single Ended, LVD or HVD bus type somewhere in its menus. For Adaptec, in their model number they have the bus type encoded. 2940 = Single Ended, 2944 = HVD, 29160 = LVD Ultra160, 29320 = LVD Ultra320. For the Device you are using, I did a search for the data sheet for the device and found out that it claimed to be Ultra160 device (It automatically has to be LVD capable to do Ultra160). >>The other thing to check is if the 2940UW BIOS has been forced to the >>slower rate. >> >> > >The card is branded Siemens-Nixdorf, I don't see the usual Ctrl-A >prompt. I've seen older SNI controllers, all settings were accessible >from the main BIOS. So, I guess I can't check this. > > > vendors in providing unique versions of firmware for their unique system configurations. It is possible that the card you are using has differences in its firmware from the generic Adaptec 2940UW such that all bets are off in its operation. It may be possible to download the generic firmware for that Controller to see if it then can operate as a generic controller, though that part would be beyond my giving advice. The card you have may be set to max out at Fast operation in the firmware for that OEM card. The vendor may have the card factory set for the type of drives that the system they were selling was using, which may have not been Ultra speed capable. >That said, when the driver is loaded, it states the USER setting of 40 >MB/s. Justin said that this is determined by EEPROM and /proc. I haven't >set anything in /proc, so I guess we can rule out the BIOS setting >limitation. > > > I agree with Justin's assessment. The OEM "Siemens" Factory Settings may be causing the negotiations with the drive to not go at the Ultra speed, or in performng domain validation, it had set to go slower in order to operate without transmission errors. The 2940 physical card itself is incapable of going any faster than Ultra, and it is limited to Single Ended communications. >The "when" part above is strange. Now I don't have the drive connected, >aic7xxx is loaded, but I see neither kernel messages in log, nor >/proc/aic7xxx. I'm running Debian sid's 2.4.18-4-k7. rmmod, modprobe >don't help. At the last boot the module was also loaded, no messages, >but rmmod, modprobe did help. The red LED named "DS1" is constantly lit. >Should I throw the card away? Or could it be anything else (power >supply???)? > > > > at the drive package, the card may have been hung while trying to scan the SCSI bus, since it would have lots of signal reflections from an unterminated Single Ended bus. The hang could put it in a state of not being able to respond until restarted, or its passed a hard reset. If you intend on using that specifc OEM version 2940, unless the firmware can be forced to generic, you may be limited to whatever speed you find operates cleanly. >>Also with Single Ended, the SCSI Cable Length can be critical. >>Depending on a number of factors, going more than 2 meters may add to >>the problem. >> >> > >1.2 m, four 68-pin connectors for drives, black box at the end saying >"Foxconn" and "Ultra 320" (terminator?), the drive connected to the last >entry before the terminator ("far" from the controller). > > > > have a raid package or JBOD with a backplane. The terminator is installed at the end of the backplane? With the drive close to the end. That does not sound like it should be a problem, though there have been problems with some backplanes in the past with running at Ultra speed when using Single Ended SCSI. >>The SCA-1 Adapter may also be adding to the problem unless it is known >>to be able to work at the Ultra burst rate in Single Ended. >> >> > >Hmm, specifically SE? The adapter claims to be "Ultra4 320/m SCSI >compliant". I'm not sure what "/m" means. MHz? It doesn't have any >active components. One resistor goes from SCA-2 to the jumper named >"LED". There are also SYN, DLY, MTR, ID3-0 jumpers, none set. > > > > from LVD to Single Ended if a Single Ended device and/or HBA is connected to the bus. Thanks, there are also SCA adapters that are not made for LVD, so I had to ask the question. >>Also verify how you have the terminations. If these are the only two >>devices, the 2940UW must have its terminations turned on (automatic >>may be ok) and there should be a termination at the drive. >> >> > >The disk is the only device on the bus. If the black box is a >terminator, then I guess I don't need termination on the drive. The >drive is branded Sun and doesn't have any jumpers. I don't know whether >one of the jumpers on the adapter is for termination; I assume none. > >At the other end, the controller has three connectors, two internal and >one external. I guess it should be terminated. There are six jumper-like >pins, four named J3, and two named J4. >http://download.adaptec.com/pdfs/user_guides/p2940uw_ug.pdf doesn't >mention anything about them, but says the termination can be set in BIOS >to Auto, Enabled, or Disabled. Is there a way to check the setting? > > > > but I do not know what it would take to do that or if its possible with the Siemens OEM version card you have. >>I believe the Fujitsu MAN3184M is Ultra160 Low Voltage Differential >>(LVD), and capable of automatically going down to Single Ended >>communications from LVD communications, like almost all LVD devices, but >>its not being used to its full potential by doing so. >> >> > >Yes. > > > > >>Changing to the 29160 controller which is Low Voltage Dfferential >>(LVD) will help immensely in terms of being able to operate at a >>faster burst rate and being able to reliably support longer SCSI cable >>lengths. >> >> > >Do you think this will work properly with the adapter? I had stability >problems with a similar adapter, AHA-2940UW, and the same drive >(although that one was not claiming Ultra 320 compliancy). > > problems are most likely associated more with Ultra Speed Single Ended SCSI than with the card itself. Ultra speed communications with Single Ended devices were at the limit of the physical bus capabilities and a lot of things could go wrong with it. Cabling, S.E. termination types, etc... LVD is much more stable, especially up through Ultra160 transfer rates. The only thing you really have to watch out for is improperly made SCSI round cables. I would recommend that you swtch to an LVD version (the 29160 should be fine) of a generic Adaptec controller, since the SCA adapter, the Box that the adapter is in, and the Drive are made to be able to support it. > >With kind regards, > > I hope this helps a bit. Best Regards, Marc _______________________________________________ [hidden email] mailing list http://lists.freebsd.org/mailman/listinfo/aic7xxx To unsubscribe, send any mail to "[hidden email]" |
Re: Adaptec 2940UW with Fujitsu MAN3184M
|
|
Hello Marc,
On Thu, Mar 01, 2007 at 05:17:36PM -0800, Marc D. Brooks wrote: > >>It is possible that even though the 2940UW initially negotiates for > >>Ultra, the domain validation forces it to the slower rate due to > >>issues seen with the communications. > >Any way to see those at the Linux administrator level? > > The type of SCSI the device and HBA supports is usually in the > specifications for the device and the HBA. Some BIOS implementations may > report the Single Ended, LVD or HVD bus type somewhere in its menus. For > Adaptec, in their model number they have the bus type encoded. 2940 = > Single Ended, 2944 = HVD, 29160 = LVD Ultra160, 29320 = LVD Ultra320. > For the Device you are using, I did a search for the data sheet for the > device and found out that it claimed to be Ultra160 device (It > automatically has to be LVD capable to do Ultra160). Thanks for the explanation! But how can I find out what factors drive the bus speed to 10 MHz? > >The card is branded Siemens-Nixdorf, I don't see the usual Ctrl-A > >prompt. I've seen older SNI controllers, all settings were accessible > >from the main BIOS. So, I guess I can't check this. ... > The card you have may be set to max out at Fast operation in the > firmware for that OEM card. ... > The OEM "Siemens" Factory Settings may be causing the negotiations > with the drive to not go at the Ultra speed, or in performng domain > validation, it had set to go slower in order to operate without > transmission errors. I see User set to 40 MB/s in /proc/scsi/aic7xxx/0, so I think BIOS settings are not the limiting factor. > >The "when" part above is strange. Now I don't have the drive connected, > >aic7xxx is loaded, but I see neither kernel messages in log, nor > >/proc/aic7xxx. ... > Was the cable disconnected at the Adaptec, or at the drive package? Sorry, the file name is /proc/scsi/aic7xxx, I hadn't looked good enough. > >1.2 m, four 68-pin connectors for drives, black box at the end saying > >"Foxconn" and "Ultra 320" (terminator?), the drive connected to the last > >entry before the terminator ("far" from the controller). > > I do not have any direct experience with "Foxconn". It sounds like you > have a raid package or JBOD with a backplane. The terminator is > installed at the end of the backplane? That isn't a backplane, that's a round cable with a black box at the end. > The Adaptec can have termination set or unset through the BIOS menus, > but I do not know what it would take to do that or if its possible with > the Siemens OEM version card you have. Ah. Again. But if the controller side is not automatically terminated, would the drive function reliably? I've built OpenEmbedded on this drive without problems (4 GB after compilation, many hours on my 1.5-GHz system). A similar configuration with a bad adapter didn't last 10 min. Can I somehow measure whether the controller has the termination enabled? > The similar Adapte you mention is sill a Single Ended one. The stability > problems are most likely associated more with Ultra Speed Single Ended > SCSI than with the card itself. Ultra speed communications with Single > Ended devices were at the limit of the physical bus capabilities and a > lot of things could go wrong with it. ... > The only thing you really have to watch out for is improperly made > SCSI round cables. I would recommend that you swtch to an LVD version > (the 29160 should be fine) of a generic Adaptec controller, since the > SCA adapter, the Box that the adapter is in, and the Drive are made to > be able to support it. That is exactly the problem: Everything is made for at least SE Ultra, but the drive operates at Fast. So my concern was why it should be better with a 29160 if the current setup doesn't do Ultra in the first place. Is there a way to check the cable and the adapter? I trust the controller, and the drive works at Ultra in a Sun workstation. With kind regards, -- Baurzhan Ismagulov http://www.kz-easy.com/ _______________________________________________ [hidden email] mailing list http://lists.freebsd.org/mailman/listinfo/aic7xxx To unsubscribe, send any mail to "[hidden email]" |
Re: Adaptec 2940UW with Fujitsu MAN3184M
|
|
Hello Baurzhan,
I think the following are the two major questions you are asking: 1. Why is my setup running at Fast instead of Ultra? From the reading you supplied, It shows the initial speed to be 40 Megabytes (20 megatransfers of a wide bus), which is Ultra speed. It then in the next two lines shows 20 Megabytes (10 megatransfers of a wide bus), which is Fast speed. This looks to have happened about the time that the Adaptec controller would have completed its Domain Validation. During Domain Validation, the Adaptec controller tries communications at various speeds to determine if the initial negotiated speed is stable for use, or if the bus has to go slower due to unstable communications while testing at the negotiated speed. Domain Validation was developed due to the problems the industry was having implementing stable Single Ended SCSI based systems that worked at Ultra speed. Most likely, your system is running at fast instead of Ultra because during Domain Validation the SCSI bus communications tested out that it would not be stable running at Ultra, so the Adaptec controller negotiated communications down to Fast. (This is if its not due to an OEM configuration built into that particular HBA you are using.) There are a lot of factors that can cause this to happen: Cables, Terminators, and Drives and/or SCA adapters that were made for LVD, which can switch down to Single Ended, but the REQ/ACK timing may be less than optimal when running Single Ended, etc... Basically, if you want to try different cables, terminators, or one or more of the major components to find the problem, this can be done. OEMs spent many hours working on combinations of hardware configurations to standardize on that would be consistently stable at Ultra Speed with Single Ended devices. The SCSI Standards Committee shortened the cable lengths, Recommended using Domain Validation (to go slower if it doesn't seem stable during startup at Ultra) and specified that Active Single Ended Terminations should be used at Ultra Speed in order to pass Single Ended SCSI communications at Ultra Speed. Add in the factor that SCSI went LVD for faster speeds than Ultra, which use different (LVD) signal transceivers than Single Ended, with Multimode versions of those transceivers able to switch down to Single Ended but optimized for LVD, and the possibility of having a system switching down to Fast for stable communications increases. What I am trying to say is that with Single Ended, you can run into exactly the problem you have experienced going at Ultra, and you may have to try changing out some or all of the components of your setup to get it going at Ultra with Single Ended. 2. What can I do to make my system run faster? I would recommend that since all of the hardware (except for the controller) is LVD, capable of running at Ultra160 (which is stable) to change the 2940 controller out for an Ultra160 capable LVD controller. You will definitely get a faster speed than the SCSI Fast you are currently experiencing, or the Ultra you ar trying to get, and not have the struggles in trying to find what part(s) in your current configuration are causing the system to run slower than Ultra. I'm sorry I cannot supply a better answer which would not have the cost of lots of time trying different components, or the cost of a different controller, but when mixing and matching Single Ended SCSI hardware at Ultra speeds, you can get a string of 1000 good setups, then hit on a string of 100 problematical ones. Best Regards, Marc Single configurations for systems Baurzhan Ismagulov wrote: >Hello Marc, > >On Thu, Mar 01, 2007 at 05:17:36PM -0800, Marc D. Brooks wrote: > > >>>>It is possible that even though the 2940UW initially negotiates for >>>>Ultra, the domain validation forces it to the slower rate due to >>>>issues seen with the communications. >>>> >>>> >>>Any way to see those at the Linux administrator level? >>> >>> >>The type of SCSI the device and HBA supports is usually in the >>specifications for the device and the HBA. Some BIOS implementations may >>report the Single Ended, LVD or HVD bus type somewhere in its menus. For >>Adaptec, in their model number they have the bus type encoded. 2940 = >>Single Ended, 2944 = HVD, 29160 = LVD Ultra160, 29320 = LVD Ultra320. >>For the Device you are using, I did a search for the data sheet for the >>device and found out that it claimed to be Ultra160 device (It >>automatically has to be LVD capable to do Ultra160). >> >> > >Thanks for the explanation! But how can I find out what factors drive >the bus speed to 10 MHz? > > > > >>>The card is branded Siemens-Nixdorf, I don't see the usual Ctrl-A >>>prompt. I've seen older SNI controllers, all settings were accessible >>> >>> >>>from the main BIOS. So, I guess I can't check this. >> >> >... > > >>The card you have may be set to max out at Fast operation in the >>firmware for that OEM card. >> >> >... > > >>The OEM "Siemens" Factory Settings may be causing the negotiations >>with the drive to not go at the Ultra speed, or in performng domain >>validation, it had set to go slower in order to operate without >>transmission errors. >> >> > >I see User set to 40 MB/s in /proc/scsi/aic7xxx/0, so I think BIOS >settings are not the limiting factor. > > > > >>>The "when" part above is strange. Now I don't have the drive connected, >>>aic7xxx is loaded, but I see neither kernel messages in log, nor >>>/proc/aic7xxx. >>> >>> >... > > >>Was the cable disconnected at the Adaptec, or at the drive package? >> >> > >Sorry, the file name is /proc/scsi/aic7xxx, I hadn't looked good enough. > > > > >>>1.2 m, four 68-pin connectors for drives, black box at the end saying >>>"Foxconn" and "Ultra 320" (terminator?), the drive connected to the last >>>entry before the terminator ("far" from the controller). >>> >>> >>I do not have any direct experience with "Foxconn". It sounds like you >>have a raid package or JBOD with a backplane. The terminator is >>installed at the end of the backplane? >> >> > >That isn't a backplane, that's a round cable with a black box at the >end. > > > > >>The Adaptec can have termination set or unset through the BIOS menus, >>but I do not know what it would take to do that or if its possible with >>the Siemens OEM version card you have. >> >> > >Ah. Again. > >But if the controller side is not automatically terminated, would the >drive function reliably? I've built OpenEmbedded on this drive without >problems (4 GB after compilation, many hours on my 1.5-GHz system). A >similar configuration with a bad adapter didn't last 10 min. Can I >somehow measure whether the controller has the termination enabled? > > > > >>The similar Adapte you mention is sill a Single Ended one. The stability >>problems are most likely associated more with Ultra Speed Single Ended >>SCSI than with the card itself. Ultra speed communications with Single >>Ended devices were at the limit of the physical bus capabilities and a >>lot of things could go wrong with it. >> >> >... > > >>The only thing you really have to watch out for is improperly made >>SCSI round cables. I would recommend that you swtch to an LVD version >>(the 29160 should be fine) of a generic Adaptec controller, since the >>SCA adapter, the Box that the adapter is in, and the Drive are made to >>be able to support it. >> >> > >That is exactly the problem: Everything is made for at least SE Ultra, >but the drive operates at Fast. So my concern was why it should be >better with a 29160 if the current setup doesn't do Ultra in the first >place. Is there a way to check the cable and the adapter? I trust the >controller, and the drive works at Ultra in a Sun workstation. > > >With kind regards, > > [hidden email] mailing list http://lists.freebsd.org/mailman/listinfo/aic7xxx To unsubscribe, send any mail to "[hidden email]" |
Re: Adaptec 2940UW with Fujitsu MAN3184M
|
|
Hello Marc,
On Fri, Mar 02, 2007 at 04:38:04PM -0800, Marc D. Brooks wrote: > I think the following are the two major questions you are asking: > > 1. Why is my setup running at Fast instead of Ultra? ... > 2. What can I do to make my system run faster? Yes. > I would recommend that since all of the hardware (except for the > controller) is LVD, capable of running at Ultra160 (which is stable) to > change the 2940 controller out for an Ultra160 capable LVD controller. > You will definitely get a faster speed than the SCSI Fast you are > currently experiencing, or the Ultra you ar trying to get, and not have > the struggles in trying to find what part(s) in your current > configuration are causing the system to run slower than Ultra. > > I'm sorry I cannot supply a better answer which would not have the cost > of lots of time trying different components, or the cost of a different > controller, but when mixing and matching Single Ended SCSI hardware at > Ultra speeds, you can get a string of 1000 good setups, then hit on a > string of 100 problematical ones. Ok, I'll look for a used 29160. Thanks to all who answered! With kind regards, -- Baurzhan Ismagulov http://www.kz-easy.com/ _______________________________________________ [hidden email] mailing list http://lists.freebsd.org/mailman/listinfo/aic7xxx To unsubscribe, send any mail to "[hidden email]" |
Re: Adaptec 2940UW with Fujitsu MAN3184M
|
|
In reply to this post by Marc D. Brooks
On Fri, Mar 02, 2007 at 04:38:04PM -0800, Marc D. Brooks wrote:
> I would recommend that since all of the hardware (except for the > controller) is LVD, capable of running at Ultra160 (which is stable) to > change the 2940 controller out for an Ultra160 capable LVD controller. > You will definitely get a faster speed than the SCSI Fast you are > currently experiencing, or the Ultra you ar trying to get, and not have > the struggles in trying to find what part(s) in your current > configuration are causing the system to run slower than Ultra. I bought a used 29160, and it worked at 160 MB/s OOTB, thanks for the suggestion! I've done some tests (hdparm -tT, build the Linux kernel with ccache) and have seen that my SATA 7.2k ST3320620AS outperforms U160 10k MAN3184M by up to 30%, so I decided not to use it for the rootfs yet. I've read that SCSI should perform better with many concurrent requests. Any simple test scenario I could try? With kind regards, -- Baurzhan Ismagulov http://www.kz-easy.com/ _______________________________________________ [hidden email] mailing list http://lists.freebsd.org/mailman/listinfo/aic7xxx To unsubscribe, send any mail to "[hidden email]" |
Re: Adaptec 2940UW with Fujitsu MAN3184M
|
Baurzhan Ismagulov wrote:
> > I've done some tests (hdparm -tT, build the Linux kernel with ccache) > and have seen that my SATA 7.2k ST3320620AS outperforms U160 10k > MAN3184M by up to 30%, so I decided not to use it for the rootfs yet. Make sure you run a fair test. Most SCSI drives are shipped from the factory with the write cache set to "write-through". SATA and ATA drives almost always operate in "write-back" mode. Your 29160 BIOS should have an option to set the caching mode on a per-target basis. If you're system is backed up by a UPS or your FS can handle being lied to, write-back is fine. Otherwise it's really not a safe option. -- Justin _______________________________________________ [hidden email] mailing list http://lists.freebsd.org/mailman/listinfo/aic7xxx To unsubscribe, send any mail to "[hidden email]" |
Re: Adaptec 2940UW with Fujitsu MAN3184M
|
In reply to this post by Baurzhan Ismagulov
On Wed, 14 Mar 2007, Baurzhan Ismagulov wrote:
> I bought a used 29160, and it worked at 160 MB/s OOTB, thanks for the > suggestion! > > I've done some tests (hdparm -tT, build the Linux kernel with ccache) > and have seen that my SATA 7.2k ST3320620AS outperforms U160 10k > MAN3184M by up to 30%, so I decided not to use it for the rootfs yet. > I've read that SCSI should perform better with many concurrent requests. > Any simple test scenario I could try? Huh? You are comparing a 1999 disk drive (MAN3184M, 18Gb) to a 2005/2006-circa SATA disk drive (Seagate Barracuda 7200.10)? It's like comparing oranges to apples IMHO. Get a 2006-circa SCSI disk drive and try again, perhaps..? As a side note, my 300GB 10k7 SCSI Seagate looks about 40% faster than my ST3750640A (Barracuda 7200.10) disk drive using hdparm -t on two Dell workstations.. Regards, -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ [hidden email] They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost _______________________________________________ [hidden email] mailing list http://lists.freebsd.org/mailman/listinfo/aic7xxx To unsubscribe, send any mail to "[hidden email]" |
Re: Adaptec 2940UW with Fujitsu MAN3184M
|
|
Hello Vincent,
On Wed, Mar 14, 2007 at 02:49:24PM +0100, [hidden email] wrote: > >I've done some tests (hdparm -tT, build the Linux kernel with ccache) > >and have seen that my SATA 7.2k ST3320620AS outperforms U160 10k > >MAN3184M by up to 30%, so I decided not to use it for the rootfs yet. > >I've read that SCSI should perform better with many concurrent requests. > >Any simple test scenario I could try? > > Huh? You are comparing a 1999 disk drive (MAN3184M, 18Gb) to a > 2005/2006-circa SATA disk drive (Seagate Barracuda 7200.10)? It's like > comparing oranges to apples IMHO. Get a 2006-circa SCSI disk drive and try > again, perhaps..? The question was more "can I still find any use for my MAN3184M, or should I throw it away", rather than "what is better". That said, I'm also interested in the latter, and I would like to do see some test results before going for, say, ST336753LW. > As a side note, my 300GB 10k7 SCSI Seagate looks about 40% faster than my > ST3750640A (Barracuda 7200.10) disk drive using hdparm -t on two Dell > workstations.. This is the disk_controller-to-RAM transfer rate, what about some simple real applications? Writing / reading /proc/kcore and building the Linux kernel from ccache differed by 6 to 20% for me while hdparm -t reported 30%. With kind regards, -- Baurzhan Ismagulov http://www.kz-easy.com/ _______________________________________________ [hidden email] mailing list http://lists.freebsd.org/mailman/listinfo/aic7xxx To unsubscribe, send any mail to "[hidden email]" |
Re: Adaptec 2940UW with Fujitsu MAN3184M
|
|
In reply to this post by Justin T. Gibbs-2
Hello Justin,
On Wed, Mar 14, 2007 at 09:41:57AM -0400, Justin T. Gibbs wrote: > >I've done some tests (hdparm -tT, build the Linux kernel with ccache) > >and have seen that my SATA 7.2k ST3320620AS outperforms U160 10k > >MAN3184M by up to 30%, so I decided not to use it for the rootfs yet. > > Make sure you run a fair test. Most SCSI drives are shipped from the > factory with the write cache set to "write-through". SATA and ATA > drives almost always operate in "write-back" mode. Your 29160 BIOS > should have an option to set the caching mode on a per-target basis. I've enabled the BIOS and set the write-back mode on MAN3184M, no difference in writing /proc/kcore. > If you're system is backed up by a UPS or your FS can handle being > lied to, write-back is fine. Otherwise it's really not a safe option. Oh. Does ext3 handle that? Or do you know filesystems that do? I've googled for that, but the links I've looked at refer to ext3's writeback mode. With kind regards, -- Baurzhan Ismagulov http://www.kz-easy.com/ _______________________________________________ [hidden email] mailing list http://lists.freebsd.org/mailman/listinfo/aic7xxx To unsubscribe, send any mail to "[hidden email]" |
Re: Adaptec 2940UW with Fujitsu MAN3184M
|
|
Hello Edward,
On Wed, Mar 14, 2007 at 12:17:22PM -0500, Edward H. Welbon wrote: > EXT3 and XFS deal with write-back crashes pretty well. XFS is faster > (especially delete) and some argue more robust. Thanks for the info! Do you have any links about that? With kind regards, -- Baurzhan Ismagulov http://www.kz-easy.com/ _______________________________________________ [hidden email] mailing list http://lists.freebsd.org/mailman/listinfo/aic7xxx To unsubscribe, send any mail to "[hidden email]" |
Re: Adaptec 2940UW with Fujitsu MAN3184M
|
|
On Thu, Mar 15, 2007 at 12:05:23AM -0500, Edward H. Welbon wrote:
> http://www.debian-administration.org/articles/388 is one place to start, > there are other links on said page. I find XFS to be durable to crashes. > I have an areca raid card with out a battery for the raid card ram on a > write back raid. Now and again the system has been crashed and powered > off and each time XFS recovered everything that had been journaled. > YMMV. I am getting a battery for the raid ram, but I've done with out > for some time. I highly recommend areca btw. Top notch for less than top > dollar. Cheers. Thanks! With kind regards, -- Baurzhan Ismagulov http://www.kz-easy.com/ _______________________________________________ [hidden email] mailing list http://lists.freebsd.org/mailman/listinfo/aic7xxx To unsubscribe, send any mail to "[hidden email]" |
|
In reply to this post by Baurzhan Ismagulov
I am having the same problem as you. However, a useful info may be that the issue is related to the Linux kernel version I am using. There is a corresponding report at:
http://bugzilla.kernel.org/show_bug.cgi?id=5808 . It is possible to recover FAST-20 mode by downgrading to kernel 2.6.12. For 2.6.13 and later, only FAST-10 mode is negociated. One can verify this on the fly by using a live CD like the one of Ubuntu 5.10 for which the full speed is achieved. |
Re: Adaptec 2940UW with Fujitsu MAN3184M
|
|
Hello acheron,
On Wed, May 23, 2007 at 09:39:29AM -0700, acheron wrote: > I am having the same problem as you. However, a useful info may be that the > issue is related to the Linux kernel version I am using. There is a > corresponding report at: > > http://bugzilla.kernel.org/show_bug.cgi?id=5808 . > > It is possible to recover FAST-20 mode by downgrading to kernel 2.6.12. For > 2.6.13 and later, only FAST-10 mode is negociated. One can verify this on > the fly by using a live CD like the one of Ubuntu 5.10 for which the full > speed is achieved. Hmm, interesting, thanks for the info. With kind regards, -- Baurzhan Ismagulov http://www.kz-easy.com/ _______________________________________________ [hidden email] mailing list http://lists.freebsd.org/mailman/listinfo/aic7xxx To unsubscribe, send any mail to "[hidden email]" |
«
Return to aic7xxx
|
1 view|%1 views
Loading...
| Powered by Nabble | Edit this page |