Hello,I have a failed drive on my RAID1+0 array with a HP P420 RAID Card.The drives are mostly Crucial m500 480GB SSDs. There were 2 Crucial m550 512GB SSDs that has been put into when 2 m500 480gb failed. And one of those 512GB SSDs failed now and I'm trying to replace it with a new m500 480GB.I've replaced the drive with a brand new drive and it didn't start the rebuilding process. Instead it kept the drive as failed in the output and with the amber led.
Dec 08, 2014 Which tools are likely to be useful to reset SMART? Is it likely that SMART will dumbly spent all reallocation space reallocating again the bad blocks from the non-partitioned space, leaving the partitioned space vulnerable to bad blocks once again? Thanks for your consideration and input. P.S.: smartctl -a /dev/sda output. How to Reset Service Wrench Icon Light Smart Car fortwo: Turn ignition to ON position. Press the left side TRIP button on the left side of the speedometer twice within the first 4 seconds. Switch off the ignition within 10 seconds. Press and hold the TRIP button again and switch the ignition to ON position.
I've changed it again with another drive. Same thing again.What may be the cause? Is it because I'm trying to replace a 512GB drive with a 480GB drive? Although, it doesn't make much sense since the array drives are set as 480GB once it was built originally.Here is the full output from hpacucli. If you haven't powered the server off it appears this may cure the problem in some other online cases. Next re-seat power and data cables in the server.Did the last drive replacement complete a rebuild?Is the the same slot you have 2x bad drives from?
The connector didn't break off did it?Have you verified the drive is dead by testing it on a PC of course with tools dedicated to testing? Possible it is the enclosure, data cables, power supply, or controller flaking out.IMO size is not the issue as the extra space is just wasted. Replaced lots of RAID drives with larger drives with no issues. Qwerty80486 wrote:If you haven't powered the server off it appears this may cure the problem in some other online cases.
Next re-seat power and data cables in the server.Did the last drive replacement complete a rebuild?Is the the same slot you have 2x bad drives from? The connector didn't break off did it?Have you verified the drive is dead by testing it on a PC of course with tools dedicated to testing? Possible it is the enclosure, data cables, power supply, or controller flaking out.IMO size is not the issue as the extra space is just wasted. Replaced lots of RAID drives with larger drives with no issues.It is the same slot. The previous drive didn't initiate the rebuild, either.The drives are verified on other servers and appear to be working.Caur wrote:I have never seen an array rebuild with a smaller drive than what the array is configured for.
Is that even possible?The array is built with 480GB drives. The drive that marked as failed was a 512GB drive. I've tried to replace it with a 480GB drive but it didn't work.-Now, I've shipped 512GB drives there and installed one of them to the same slot and nothing happened again.Still same.
I'm hoping that a kind soul with more expertise than I might throw me an assist. I'm no expert, but believe I should be able to manage resetting SMART on a Seagate ST2000LM007. I picked up a PL2303HX USB-TTL RS232 Arduino off eBay, but am having difficulties with getting terminal T. Based on my research, the pin config for the diag port should be: Pwr GND TX RX - SATA - PWR -So, I've connected my pins accordingly; using 38400 8N1 None with Realterm, and have played around with multiple pin configs.
I just can't seem to get to T no matter what. About the best I'll manage is either a locked session that is unresponsive, or depending on the config it will spit a bunch of garbled data (in ascii- hex will display characters).Wondering if my usb-ttl is junk, or rather inappropriate for this job.Prolific USB-TTL pin scheme- Red: + 5V- Black: Ground- Green: RXD- White: TXDHoping someone can steer me towards the right config with the hardware I have, or steer me towards what I need to purchase. I have 2 Seagate drives (ST2000LM003 & ST2000LM007) that I'd like to reset SMART on.
I've tried to include the pertinent information in the pics.Thank you so much! Your feedback is super helpful; thank you.
The issue with this specific drive is that I lost power during a Secure ATA erase. From that point, SMART was tripped and the drive rendered inoperable. However, I believe the drive itself is perfectly fine.I attempted using the Seachest cli tools to no avail, so this seemed the next logical step. I'll start by picking up a CP2104 chip. However, with respect to the 'terminal access blocked' issue. Is there workaround?Thanks again for your assistance.
Thank you very much for the feedback. I'll assume that something like this will do the trick.I feel like an idiot, but to save time- I was performing a Secure ATA Erase on 2 drives at the same time; a ST2000LM003 & ST2000LM007 and suffered a power outage. Both drives were rendered inoperable, but should be mechanically perfectly fine, other than the SMART was tripped on both.
I'm not in 'the business', but am technically inclined. I've just have never delved this deep into drive recovery stuff before.
![Smart Smart](/uploads/1/2/5/6/125638916/139086255.jpg)
I appreciate your willing to point me in the right direction.Are there any known workarounds for the 'block' on terminal for these Rosewood drives?Thanks again. The problem you see probably is not from smart, but due to the drives being locked with password before the secure erase and that password not having been removed.Try to redo the erase, with the same tool and password as before. Maybe this will solve your problem.It will not.Thos drives are junk and i would say that Security Erase did attempt to write zeros to all sectors finding a huge amount of bad ones that were re-located (swapped by spares).If you loose power during security erase you will send up with a drive locked with ATA password and no user space access - LBA - but there is no reason for S.M.A.R.T. To show BAD attributes.If you have some attributes as re-location sector count or pending sectors, etc, then the drive have severe media damage (as expected on those models) and you should bin the drive, buy a new one and be happy that you don't need the data!Do not ever again use a Seagate F3 arch drive as well.
![Smart Ncq Reset Smart Ncq Reset](/uploads/1/2/5/6/125638916/632709131.jpg)
Okay, so please bear in mind that I am unable to get to terminal on this drive, at the moment. The tools that I am using are within the SeaChest git, which offers Windows and Linux binaries. If there is a better way for me to access SMART logs. I am all ears.
But, here is what I have (and the process that I've used).A couple important aspects to take note of. Please notice: 'SMART Status: Tripped & ATA Security Information: Supported.'
I believe that I've been able to clear the password from the previous Secure ATA Erase, in a previous attempt. As I understand it, if it were locked due to password. That field would say: 'Supported, Frozen'.Notes: BIOS recognizes drive. Linux & Windows recognize the drive.
In Windows, Disk Management will prompt me to Initialize the Disk, and to give it a partition table MBR/GPT. Selecting either options gives me 'Data Error (cylic redundancy check)'. If I dd directly to write the fail, bombs out with failures.Here is a log. I don't know this forum didn't post my previous reply, but here are the logs that I am able to produce. Bear in mind that I don't have terminal, so the best tool that I am aware of is the Seachest git.Notes: BIOS recognizes drive. Windows recognizes drive. Disk Management prompts me to Initialize the Disk, and to give it a partition table MBR/GPT.
Selecting either options gives me 'Data Error (cylic redundancy check)'.A couple important aspects to take note of. Please notice: 'SMART Status: Tripped & ATA Security Information: Supported.' I believe that I've been able to clear the password with the Secure ATA Erase, in a previous attempt. As I understand it, if it were locked due to password. That field would say: 'Supported, Frozen'.