Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Topics - adridolf

Pages: [1]
News and Updates / Forum Mails sent after ages
« on: November 20, 2017, 04:08:58 am »
I just received a bunch of mails notifying me about topic replies from between May and August. (I never got them in time before, but did not report so far.)

Maybe check how those mails are sent, because right now it is just annoying.

Despite the MTBF (mean time between failure), drive specs typically include a value for the unrecoverable read error per bits read (URE). For consumer drives this frequently is 1 per 10^14, so 1 error per 100 Terabits or 12.5 Terabytes.

It is stressed in various threads that for traditional RAID-5 arrays, this would mean a very high failure rate if those values were actually accurate.

In the context of FlexRaid/t-Raid, I wonder how this is connected to Scheduled Verify or Verify/Sync tasks. I personally use a scheduled Verify/Sync for 100 GB daily (with 4 TB PPU). So I will scan about 3 TB per month and 36 TB per year and disk, meaning three unrecoverable read errors per disk per year (without proper stochastical treatment, as this is not relevant for the question).

So my questions in this context are:
1. How does this affect the health of the whole array? Will I get silent corruption of a few individual bits here or more serious effects or nothing at all?
2. Would it be more reasonable to just do a scheduled Verify or Verify+ task, so I can detect "errors", but do not automatically propagate wrong bits?
3. One idea would be to stop the scheduled tasks to go around the URE problem. I however suppose the benefits of this "health check" very much outweigh the disadvantage of an occasional URE?

tRAID on Windows / NTFS allocation size
« on: February 13, 2017, 05:57:45 am »
Is there any restriction/interference between the NTFS allocation size chosen during formatting and tRaid or can it just be set based on the considerations you would take for normal drives?

I set up a new tRaid array and want to plan my formatting accordingly.

General Discussion / Option to edit scheduled job
« on: February 07, 2017, 06:44:16 am »
Is there a possibility to edit a scheduled job (e.g. change start time from 15:00 to 16:00)?

At the moment, I delete the old job and create a new one...

tRAID on Windows / How do I recognise a failed disk
« on: February 06, 2017, 06:52:22 am »
If I understood the mechanism correctly, a disk is not "failed" in tRaid if it has been just disconnected or is malfunctioning.

I also understand that this is considered a feature, so disks are not failed only because of connectivity issues etc.

As a consequence, the question arises how I can actually find out about a malfunctioning drive in a tRaid array in time to quickly recover it? Am I supposed to extract this by SMART values etc. or is there a mechanism implemented in FlexRaid?

Note that this is not supposed to criticize the way tRaid handles this, but to find out about a (best-practice) solution for quickly identifying failed drives.

News and Updates / No notification mails sent
« on: February 03, 2017, 10:13:46 am »
I do not get any notification e-mails if threads are updated or even if I receive a private message. I've checked and enabled all related settings in my profile, but still don't receive anything. Does anyone have the same issue or is there a hidden setting I do not know about?

I attached a screenshot of the settings, I also get the correct list of topic notifications in the menu. E-mail address is correct...

Pages: [1]