Potential Samsung F4 issues.

[pras1011]

Is there a way to update the hdd by using a laptop with a estata pci express card and a esata hdd exclosure?

[slaveunit]

Just curious...how does a laptop have a pci express card in it?

[pras1011]

Its has a expresscard slot on the side of the laptop.

[wsume99]

The EARS and the Seagate LP have potential problems in how long they will work for due to their load/unload cycle problems.

Why not just disable the IDLE3 timer on the WD drives and then you won't have to worry about load cycles?  Am I missing something?

[slaveunit]

IF you can boot into DOS from a flash drive and it will see the esata port, I dont see why it wouldnt work.

[pras1011]

I have DOS 7.1 on my usb drive. I run the patch but it doesn't see the esata express card.

[Joe L.]

I have DOS 7.1 on my usb drive. I run the patch but it doesn't see the esata express card.

MS-Dos had no SATA drivers... never did.  Even Win-XP did not.    Not sure what you should do but I can guarantee you'll be in good company.  I think you'll need to be on a motherboard where the drive can be set to emulate an IDE drive.  (Known a legacy mode in some BIOS)

[slaveunit]

I have a plain jane Gigabyte MB with the drive connected via sata.  I didnt look to see if I had any legacy enabled but the update worked fine on it.

 

Pras1011 did you try the page I linked for the DOS system files?  Once you build that you just copy the samsung file on it and boot from the flash then run the samsung exe file.

[pras1011]

I haven't tried it. I am going to get a PC to be on the safe side.

 

I just wanted to let you know the firmware version will not change after flashing. The version did not change because

 

it takes a long time to get customers approval for change notices. It would have taken over a month before we could

 

have released the fix.

 

Please make sure that for each drive the flashing process is successful…

 

We have already implemented the fix to our production. So drives available on the market in 2 months time will have the

 

fix applied. To be sure check the manufacturing date on the drives, January 2011 would be safer…

[LaurentHU]

> Ouch I have 2 of them running!!!!

Well, gotta patch firmware then... What's the best way of doing it? Anyboddy having experience? Shall I just unplug all disks but the Samsung ones and update them 1 by 1 in the UnRaid server?

[slaveunit]

> Ouch I have 2 of them running!!!!

Well, gotta patch firmware then... What's the best way of doing it? Anyboddy having experience? Shall I just unplug all disks but the Samsung ones and update them 1 by 1 in the UnRaid server?

 

You could try reading one page back where I posted "I think the easiest way to do this is".  

 

I didn't check to see if my data was erased from the drive.  Before I updated it I took it out of the array and started the array without it.  That way it would rebuild after the update.  I did only have one data drive that was a samsung F4.

[LaurentHU]

Yes I saw your post, and actually managed to create a bootable DOS USB stick following the instructions - HP program and Dos files -, and did put the firmware file on it. Tried it on a test machine, works very well - it scanned all disks and did nothing because the Samsung drives were not there.

I'm now doing a parity check on my Unraid rig, read only, to assess damages before doing firmware upgrade... Maybe OK since I had NCQ disabled? The different websites are not clear whether that really saves you or not... So far 11%, and 0 faults. Since I have 2 drives, not much I can do anyway!

[LaurentHU]

Ok, so results of the day :

* 0 faults on parity check - makes me think the issue is not that common in "normal" Unraid situation. I have the disk since almost a month, they are more than 50% full, and I do write a lot to them - I have a cache drive but I also do update on existing files. I'm using unmenu and I look at it frequently.

* firmware update worked like a charm - just had to boot on DOS USB stick on my rig, run the Samsung program, and it scanned all disks, found the 2 faulty disks and patched them (with confirmation). Since then rig running, nothing to declare.

So maybe I was damn lucky, maybe not all drives were faulty, but in my case much more noise than harm... Wish it's the same for all folks there.

[Giraffeninja]

The issue doesn't occur in normal unRaid use, it could only show up if you were running add ons on the server (like unmenu) and only if the drives were connected using ahci.

[LaurentHU]

Precisely I am running unmenu and the drives use AHCI...

[BRiT]

* firmware update worked like a charm - just had to boot on DOS USB stick on my rig, run the Samsung program, and it scanned all disks, found the 2 faulty disks and patched them (with confirmation). Since then rig running, nothing to declare.

 

How do you know the firmware update worked? You don't!

 

That's right, no one knows if the firmware flash was successful unless they attempt to reproduce the bug that corrupts your data.

[LaurentHU]

Well yes you're right you can't know for sure - but the firmware update program tells it to you very clearly. So I agree it's a big issue they didn't update the firmware version, but again when you run the program it tells you very clearly that the update was successful.

 

[prostuff1]

Well yes you're right you can't know for sure - but the firmware update program tells it to you very clearly. So I agree it's a big issue they didn't update the firmware version, but again when you run the program it tells you very clearly that the update was successful.

 

 

True, but do you really think it is that hard to print "success" on the screen no matter what happens.

 

They should have changed the versioning number, plain and simple.

[BRiT]

That's the exact same link included in the first post and what started this entire thread.

[Copey]

Mmm, I think I may have had a similar problem with Samsung HD103SI drives 

 

Recently started using unmenu 'mymain' to check write/transfer speeds whenever I send data to my server, and (by coincidence) I've gone thro' a bunch of HD103SI drives!

I had put it down to a faulty batch of drives, but now I'm not so sure, I'm not blaming umenu... but simply pointing out the link.

 

AFAIK, unmenu seems to work ok with other HDD brands???

 

 

 

[Joe L.]

Mmm, I think I may have had a similar problem with Samsung HD103SI drives 

 

Recently started using unmenu 'mymain' to check write/transfer speeds whenever I send data to my server, and (by coincidence) I've gone thro' a bunch of HD103SI drives!

I had put it down to a faulty batch of drives, but now I'm not so sure, I'm not blaming umenu... but simply pointing out the link.

 

AFAIK, unmenu seems to work ok with other HDD brands???

unMENU only reports on the disk statistics stored by OS and by bwm-ng.  So... don't blame the messenger. 

[BRiT]

If there is a problem with the Samsung HD103SI drives too put the blame where it belongs, with Samsung. It does not belong with the standardized OS or drive tools. It falls squarely on the drive manufacturer.

[Copey]

Hi Joe L.

 

Apologies, not shooting the messenger at all.

I'm not a Linux guru by any measure, so please forgive my ignorance 

 

I had experienced 100% reliability using 10 of these drives until I started using unmenu.

I'm assuming that I am calling the 'identify' command whilst I'm fumbling around unmenu and triggering the fault?

 

I have already contacted Samsung to advise them of this issue... I just hope that they are prepared to do something about it!

 

 

Cheers

[Joe L.]

Hi Joe L.

 

Apologies, not shooting the messenger at all.

I'm not a Linux guru by any measure, so please forgive my ignorance 

 

I had experienced 100% reliability using 10 of these drives until I started using unmenu.

I'm assuming that I am calling the 'identify' command whilst I'm fumbling around unmenu and triggering the fault?

 

I have already contacted Samsung to advise them of this issue... I just hope that they are prepared to do something about it!

 

Cheers

They've already issues an upgrade for their firmware that causes incomplete "writes" of blocks in their internal queue.    That is a completely different issue than the one you are reporting of slow performance.  That could be a driver issue, or a disk controller issue, or a cabling issue, or even a power supply issue in addition to firmware on the specific disk not working well with the default buffer sizes as set by the unRAID disk driver.

 

As far as the disks working well under windows... how did you know?  did you run the same types of tests and timings?  I'm certain many people "think" their disks are working well when they are no set up for optimum performance compared to their competitors and I know I've never run my window's disk through the same measurements for performance.

[furymaster]

Hi,

 

as a frequent reader of this forum I realise that the F4 is having issues with unRaid.

 

As I am currently planing some kind of "Movie Monster" I've been wondering if there are some different versions of the Samsung F4.

 

Here is what I've found out:

 

http://www.hardwareluxx.de/community/f15/samsung-f1-f2-f3-f4-festplatten-4-a-752441.html

 

HD204UI - "512B"

HD204UI/Z4 - "4K Sector"

HD204UI/UZ4 - "4K Sector with 512B Emulation"

 

Question is - which of these three is the one with the "performance issue"? Are they all the same? Or is one of them useful with the current version of unRaid?

 

Thank you,

 

Herbert

 

P.S.: Having some 6000+ DVDs laying around, my options are

a) recruiting some space from my wife (bad option, very low WAF)

b) building an unRaid Server (better option, WAF yet to be tested)

 

P.P.S.: Excuse my bad english. Learned english in school, but I have very seldom contact to native speakers...