[Irinikus]

<QUOTE author="gijoe77" post_id="2406" time="1534265954" user_id="243">

tablet is sort of a pain, I would get a smaller one than the big massive one. I took the fibre channel cards out and put SATA cards in, I have 4 SATA drives where the CDROM drive was and I have an external SATA JBOD case with 8x2TB SATA drives. I had planned on making the 4 sata drives that are in my CDROM drive area my Framstore (It does something like 200+ MB/s r/w, and the 8 disks on the jbob a very very fast workspace/scratch drive with a backup of all my stuff(perhaps even perhaps link /usr to it - but I just haven't gotten around to it). Too many other interesting things to do with these machines taking up all my time

A smaller tablet it will be then.

Your idea of going with SATA cards is the better option, but I will go with the fibre channel array, just for the sake of keeping the system period specific. (I know it's crazy though!)

Thanks for the good advice, I really appreciate it!

[jan-jaap]

<QUOTE author="gijoe77" post_id="2403" time="1534264208" user_id="243">

If you want to actually capture video via HD-SDI or SD-SDI then you would need a VBOB w DM5 card, a DM3 card in the Tezro (actually I forget the DM numbers but you get the idea)

I think you need the DM3 in the Tezro which you connect to the VBOB with a pair of LVDS cables. This gives you SDI/HD-SDI input/output.

With an optional DCD on the V12 you can connect to an optional DM5 in the VBOB with a pair of DVI cables, this gives you HD-GVO capabilities (graphics-to-video). In that case, the display is connected to the third DVI port on the DM5. At some point SGI discontinued the DCD and you needed an external DVI splitter box, like in the attached picture. The whole thing is described here: <URL url="http://download.autodesk.com/us/systemdocs/pdf/EffectsEditing2007_Hardware_Setup_SGITezro-Octane2.pdf"><LINK_TEXT text="http://download.autodesk.com/us/systemd ... ctane2.pdf">http://download.autodesk.com/us/systemdocs/pdf/EffectsEditing2007_Hardware_Setup_SGITezro-Octane2.pdf</LINK_TEXT>

I, too, have a "Discreet config" Tezro. I got most of the periperals (DM5, LVDS cables etc) with the system. I added the FC storage, Lucid ADA8824 and a Wacom tablet myself. I even have the Discreet dongle and the original system disk

The VBOB is a noisy bugger though so I don't want to have it on unless I need it, so I also don't feed the display output via the DM5. I normally use another, non-Discreet system disk so I don't muck up the Discreet setup.

[sgi-enthusiast]

Hi,
I have a Flame 2008 setup on a Tezro and recently changed out the older 73Gb drives in a tray for 300Gb versions. But no matter how I configure them in sw_config I cannot save the configuration as this error message appears "Saving device information failed. loctl SWR SET CONFIG failed for partition 0 (/dev/swr00) Disk 5 (adapter 4, target 17) appears twice in the data disk list for partition id 0"
If I change the parity drive and spare drive in any combination I get the same error but with a different disk, adapter and target numbers. Every combination produces the same error message and I can't save any configuration.
I checked the steps in the 'Filesystem and Networking Guide' but the error persists.
Out of frustration I put a new disk into the Tezro and built a system from scratch but even then the same error message occurs and I cannot save the configuration.
Has anyone encountered this error before and found a fix?
Regards

[gijoe77]

if you have two same model drives and they are on the same scsi 0 chain, and disk 1 is os disk, disk2 is stonefs disk, sw_config has a problem. put old 73gb os disk 1 in, try to make new 300gb disk2 stonefs. after the stonefs is made, put 300gb os disk back in

[sgi-enthusiast]

Thanks gijoe77 I will give that a try.
Interestingly, all the 73Gb drives in the tray are identical make and model number. I thought this was a requirement otherwise sw_config complains about the drives not having the same block size meaning you have to manually alter them so they all match?
Regards

[gijoe77]

I had a post about this on nekochan, basically if the os disk is the same make/model as stone disks, but os disk comes earlier in the scsi chain, it seems to throw sw_config off.

[sgi-enthusiast]

Okay but if the drives are different then you wouldn't expect to have that clash. The Tezro boot drive (before I changed it on the weekend to a ST3300007LC which is 300Gb but different model) was a Samsung 500Gb SSD in an ACARD SCSI case. The drives in the tray are all Seagate ST3300655FC 300Gb.
How would it confuse the two?
Thanks for your assistance on this.

[Irinikus]

<QUOTE author="sgi-enthusiast" post_id="2782" time="1535011993" user_id="214">

Hi,
I have a Flame 2008 setup on a Tezro and recently changed out the older 73Gb drives in a tray for 300Gb versions. But no matter how I configure them in sw_config I cannot save the configuration as this error message appears "Saving device information failed. loctl SWR SET CONFIG failed for partition 0 (/dev/swr00) Disk 5 (adapter 4, target 17) appears twice in the data disk list for partition id 0"
If I change the parity drive and spare drive in any combination I get the same error but with a different disk, adapter and target numbers. Every combination produces the same error message and I can't save any configuration.
I checked the steps in the 'Filesystem and Networking Guide' but the error persists.
Out of frustration I put a new disk into the Tezro and built a system from scratch but even then the same error message occurs and I cannot save the configuration.
Has anyone encountered this error before and found a fix?
Regards

This is pretty much the exact same problem as I had, read the earlier posts in this thread to see how the problem was solved.

[gijoe77]

<QUOTE author="sgi-enthusiast" post_id="2789" time="1535025978" user_id="214">

Okay but if the drives are different then you wouldn't expect to have that clash. The Tezro boot drive (before I changed it on the weekend to a ST3300007LC which is 300Gb but different model) was a Samsung 500Gb SSD in an ACARD SCSI case. The drives in the tray are all Seagate ST3300655FC 300Gb.
How would it confuse the two?
Thanks for your assistance on this.

It sounds like you might be having a different problem then.

Im traveling now and can't look at my notes, but it was something like the label that s+w uses to put in the volume header of the stonefs disks just looks at the model number (and it wasn't the full model number either), it's too dumb to tell if a disk was an os disk or not, but worth trying a completely different disk if it's not too much trouble to rule out.

I tried finding my post on archive.org/nekochan where I went into more detail but couldn't find it

Edit I also lost entire arrays because S+W decided to change all the volume headers on an array, so be careful if you have an external array and half the drives are meant for a xfs filesystem and half are for stonefs. When my scsi disks 1-8 were for xfs and 9-16 we're for stonefs, stone fa just went ahead and relabeled the first disks. If I made the early disks stonefs and the later disks were xfs it worked fine

[sgi-enthusiast]

<QUOTE author="sgi-enthusiast" post_id="2789" time="1535025978" user_id="214">

Okay but if the drives are different then you wouldn't expect to have that clash. The Tezro boot drive (before I changed it on the weekend to a ST3300007LC which is 300Gb but different model) was a Samsung 500Gb SSD in an ACARD SCSI case. The drives in the tray are all Seagate ST3300655FC 300Gb.
How would it confuse the two?
Thanks for your assistance on this.

It sounds like you might be having a different problem then.

Im traveling now and can't look at my notes, but it was something like the label that s+w uses to put in the volume header of the stonefs disks just looks at the model number (and it wasn't the full model number either), it's too dumb to tell if a disk was an os disk or not, but worth trying a completely different disk if it's not too much trouble to rule out.

I tried finding my post on archive.org/nekochan where I went into more detail but couldn't find it

Edit I also lost entire arrays because S+W decided to change all the volume headers on an array, so be careful if you have an external array and half the drives are meant for a xfs filesystem and half are for stonefs.  When my scsi disks 1-8 were for xfs and 9-16 we're for stonefs, stone fa just went ahead and relabeled the first disks.  If I made the early disks stonefs and the later disks were xfs it worked fine

Hi gijoe77,
I have re-read this post and the Discreet "FSNG" to determine if I had made any errors in the steps but if I have I can't see where.
Given the 'dumbness' off sw_config I decided to remove the 300Gb boot drive I made on the last attempt and replace with a 146Gb model which is the only one in the system.
The current configuration is,
Boot drive - 146Gb SCSI
Stone Drives - 11x 300Gb FC
Aside from the CDROM drive, nothing else is connected (excluding non-disk relevant items like VBOB, ADAT and BMD video gear)
I installed IRIX from scratch (6.5.30) and reinstalled Flame. Stopped Stone+Wire and unmounted the non-existent partition.
Followed the steps in the FSNGuide to make sure and exactly the same error occurs. I cannot save any form of configuration.
Any ideas?