• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

quick disc / array failure noob question

P

philster

n00b
Joined
Mar 23, 2010
Messages
29
Hi,


Im in the process of building a home server. For the streaming and storage of movies, tv clips and music. I haven't worked out what to use freenas or unraid whichever is good at both streaming HD content and good disc failure support / redundancy (if its better to use amahi for streaming then i will)

I expect to use either zfs raid-z2 or raid 6

ok the quick question is given that my server will use the 6 on-board sata 6g connectors for storage (usb flash for O/S) and no after market raid card or hot swap with backplane how will I typically know which disc has failed in the array? any handy tips? i wont have lights and buzzers to alert me which one has gone

I know its a really dumb question but do I just write down after looking on bios which sata port my new HDD is each time i add a disc? and perhaps label the discs physically for port and name? or is that pointless as the NAS software will tell me which disc is which based on serial number perhaps ? and then should i label them or have a txt document with which HDD`s corresponds to which physical location inside the computer, any help appreciated

sorry for the lame question :)
 
At work on linux I do this by asking the drives for their serial #s. I do that with hdparm or smartctl. Then finding the bad drive by serial number. I have 100+ drives in linux software raid and I have replaced quit a few drives over the years so this works pretty well. Since failure rates have increased I have moved to putting most of my drives in hot swap bays that have lights. Then to find the bad drive I use dd to read data from the drive looking for the light if it is still working or read from the arrays if it is not to track down the bad drive.
 
drescherjm said:
At work on linux I do this by asking the drives for their serial #s. I do that with hdparm or smartctl. Then finding the bad drive by serial number. I have 100+ drives in linux software raid and I have replaced quit a few drives over the years so this works pretty well. Since failure rates have increased I have moved to putting most of my drives in hot swap bays that have lights. Then to find the bad drive I use dd to read data from the drive looking for the light if it is still working or read from the arrays if it is not to track down the bad drive.
Click to expand...

so from a home use perspective, when using the onboard 6g sata`s it would be a good idea to get the serial numbers and document them. Rather than label the fronts of the drives I could just write down on the document which physical bay out of the 6 is tied to which serial number on the HDD?

makes sense, its just I don't want to go for a raid card or a hot swap bay at home. for one there is too much fan noise and dust build up at the rear using these bays and its not going to be used extensively for multiple people at a time but it will be streaming proprietary file formats and transcoding high bit-rate content. The raid card isn't needed for my usage pattern, the cost of a decent primary card and the need for a spare raid card also puts me off.

thanks
 
BTW, I am not using HW raid cards either. I now use mostly LSI based SAS HBAs + motherboard ports.

Before the hotswap bays I used to put stickers on the drives to help with the serial number verification.

Rather than label the fronts of the drives I could just write down on the document which physical bay out of the 6 is tied to which serial number on the HDD?
Click to expand...

That would be fine as long as you keep the list up to date. Obviously that would be easy for your setup.
 
drescherjm said:
BTW, I am not using HW raid cards either. I now use mostly LSI based SAS HBAs + motherboard ports.

Before the hotswap bays I used to put stickers on the drives to help with the serial number verification.



That would be fine as long as you keep the list up to date. Obviously that would be easy for your setup.
Click to expand...

thanks for the advice. On second thought with stickers it wouldn't be much quicker to see if a disc had failed. If I received a snmp trap log or was just browsing to the server then i would already have to be on a computer via weblogin which hopefully would have access to the txt document locally

again many thanks
 
I would make sure you use serial numbers for whatever method versus device names. Since I believe device names can change. At least they do this on linux. I do not have that much experience on non linux zfs setups so I am not sure if that too is a problem.
 
You must log in or register to reply here.
Back
Top