[Murray North]

Sorry Chris, my post wasn’t a stab at you. I think somewhere earlier in the thread someone had said that it was impossible and the worst idea ever, so it was more a nod to that. At any rate, you clearly have more drive and RAID knowledge than me, and i’m sure everything you say is true. But all I can say is that for better or worse, we have a few 100TBs of green drive raids, and we follow every precaution we can, regular verifies and the rest of it, and to date we have lost no data. I haven’t had to deal with random hangs, and rebuilds go without issue. I might just be very lucky, but I am just reporting what has happened to me.
Thanks for the tips though 🙂

[Murray North]

Thanks for the input Chris. We addressed any cabling issued very early in the piece and had high grade cat6 installed to eliminate the chance of this being a problem.
Just to sign off on this whole saga, we discovered that the problem wasn’t just random dropped frames, it was specifically FCP having problems as soon as it had to access media from a different raid array from the one it was looking at currently. That is, if you were playing back media from only raid A or only raid B, it would be fine, but if you had a sequence with media on raid A, and the playhead approached media from drive B, final cut would drop frames and potentially not recover for 5 or 10 seconds at worst.
We’ve installed a new raid box, media managed all the important media onto it, and now the editors work exclusively from it, and it works capably. To anyone who says green drives can’t work in a raid, and work reliably and well, respectfully, you are wrong. That isn’t to say that we aren’t buying red drives from here on in, but whatever the problem is, and it may still be a green drive thing, it can still work in certain circumstances.
Thanks everyone for their help though, and I hope that this can be of help to someone else. Feel free to message me if you are having similar problems.

[Murray North]

Thanks for the responses.

I didn’t want to get too bogged in the network setup as my hunch is that it is a problem with the drives. Once I have eliminated the drives as a problem, if any of you have the patience I will outline my setup and see if I can find any problems there, but at this stage I don’t want to waste anyones time.

Thanks for those suggestions Frank, after reading extensively through these forums however, I feel as though I have ticked all those boxes unfortunately.

For your interest the switch is a HP PROCURVE 2910AL-48G SWITCH 48-port 10/ 100/ 1000 basic L3 fixed port switch.

Currently, however, I am just keen on working out whether the drives are providing and insurmountable hurdle to my shared storage system here, and whether maybe wdidle3 or something else can magically save the day.

Will get back soon, thanks all 🙂

[Murray North]

Oh and the switch is
HP PROCURVE 2910AL-48G SWITCH 48-port 10/ 100/ 1000 basic L3 fixed port switch

[Murray North]

Haha I love all these responses! Bob I am suffering and will likely continue to, but that’s all part of the fun! Okay so, whilst in theory I understand that perhaps my green raid should be falling to pieces and failing, it and others like it are kicking ass at backing up our editshares and storing raw camera data etc.
The drive fail rate is unremarkable, though higher certianly than RE drives, and the RAIDs do not degrade every 10 minutes from drives spinning down. We have ~300TB of this kind of storage and I’d go so far as to recommend the ‘ol green drives for this purpose.

However that is neither here nor there as this particular raid is being used for editing and seems to be causing the problem.

Running this wdidle on 48 odd drives one by one on some DOS system is a daunting task, and I hope to avoid it, but yes Alex I plan on trying to find a way to test this first, although I don’t think I have the hardware to do so really.

Does anyone have any input as to whether my symptoms can be found on RE drive systems, or perhaps if Red or Black drives are up to the task, or if THEY exhibit similar symptoms as well?

Thanks all 🙂

[Murray North]

I was having problems with layers importing, though under slightly different circumstances.
I was seriously about to have a coronary, but then I figured that 16 bit color psd’s don’t seem to work; you have to change it down to 8 bit (Image -> Mode -> 8 bits / channel)
Hope this helps someone.

[Murray North]

Thanks John, i’ll keep the database going then.

[Murray North]

Thanks Brendan!
Yeah that is all 6.0 and above as you say, but handy to know for when we get there.
It’s true that in an avid 6 environment the DX works with the decklink card, but unfortunately back in 5 land, you need to have the computer running in 32bit kernel for avid to see the DX at all, which means that DaVinci 9 is a no go.
Anyhow, fun and games.

[Murray North]

Okay. I’ve come up with a resolution which i’m reasonably happy with.

The blackmagic desktop video drivers do something which stops avid from firing up the aja IO express drivers as they are supposed to.
After deleting random bits of the blackmagic drivers, i discovered that the offending part of the driver was
~/Library/Applications Support/Blackmagic Design/Blackmagic Decklink/DeckLinkAPI.bundle

If you delete or move this file and THEN boot avid, avid will recognise your IO express as it should.
You can add the file back at any time to bring back functionality to your blackmagic IO. What’s nice here is that you can remove the decklinkAPI before you boot avid, return it once avid has booted, and then boot up davinci and get the party started.

The uh, grade and cut party, I guess.

Hope this helps someone…

[Murray North]

Sorry I meant has anyone got davinci working with blackmagic hardware etc. running concurrently with a Avid on a Mojo DX?