Re: [RFC] Btrfs device and pool management (wip)

2015-12-08 Thread Christoph Anton Mitterer
On Mon, 2015-11-30 at 13:17 -0700, Chris Murphy wrote: > On Mon, Nov 30, 2015 at 7:51 AM, Austin S Hemmelgarn > wrote: > > > General thoughts on this: > > 1. If there's a write error, we fail unconditionally right now.  It > > would be > > nice to have a configurable number

Re: [RFC] Btrfs device and pool management (wip)

2015-12-02 Thread Goffredo Baroncelli
On 2015-12-02 00:43, Qu Wenruo wrote: [...] > > And block layer provides its own listen interface, reporting errors > like ATA error. Could you point me to this kind of interface -- gpg @keyserver.linux.it: Goffredo Baroncelli Key fingerprint BBF5 1610 0B64 DAC6 5F7D 17B2 0EDA 9B37 8B82

Re: [RFC] Btrfs device and pool management (wip)

2015-12-02 Thread Qu Wenruo
On 12/03/2015 03:07 AM, Goffredo Baroncelli wrote: On 2015-12-02 00:43, Qu Wenruo wrote: [...] And block layer provides its own listen interface, reporting errors like ATA error. Could you point me to this kind of interface Not yet, and that's the problem... Thanks, Qu -- To

Re: [RFC] Btrfs device and pool management (wip)

2015-12-01 Thread Goffredo Baroncelli
On 2015-11-30 13:43, Qu Wenruo wrote: > > > On 11/30/2015 03:59 PM, Anand Jain wrote: >> (fixed alignment) >> >> [...] > > I'm overall OK with your *current* hot-spare implement. > It's quite small and straightforward. > Just hope some more more easy-to-implement features, like hot-remove

Re: [RFC] Btrfs device and pool management (wip)

2015-12-01 Thread Brendan Hide
On 11/30/2015 11:09 PM, Chris Murphy wrote: On Mon, Nov 30, 2015 at 1:37 PM, Austin S Hemmelgarn wrote: I've had multiple cases of disks that got one write error then were fine for more than a year before any further issues. My thought is add an option to retry that

Re: [RFC] Btrfs device and pool management (wip)

2015-12-01 Thread Brendan Hide
On 12/1/2015 12:05 PM, Brendan Hide wrote: On 11/30/2015 11:09 PM, Chris Murphy wrote: On Mon, Nov 30, 2015 at 1:37 PM, Austin S Hemmelgarn wrote: I've had multiple cases of disks that got one write error then were fine for more than a year before any further issues.

Re: [RFC] Btrfs device and pool management (wip)

2015-11-30 Thread Chris Murphy
On Mon, Nov 30, 2015 at 1:37 PM, Austin S Hemmelgarn wrote: > I've had multiple cases of disks that got one write error then were fine for > more than a year before any further issues. My thought is add an option to > retry that single write after some short delay (1-2s

Re: [RFC] Btrfs device and pool management (wip)

2015-11-30 Thread Chris Murphy
On Mon, Nov 30, 2015 at 7:51 AM, Austin S Hemmelgarn wrote: > General thoughts on this: > 1. If there's a write error, we fail unconditionally right now. It would be > nice to have a configurable number of retries before failing. I'm unconvinced. I pretty much immediately

Re: [RFC] Btrfs device and pool management (wip)

2015-11-30 Thread Austin S Hemmelgarn
On 2015-11-30 15:17, Chris Murphy wrote: On Mon, Nov 30, 2015 at 7:51 AM, Austin S Hemmelgarn wrote: General thoughts on this: 1. If there's a write error, we fail unconditionally right now. It would be nice to have a configurable number of retries before failing. I'm

Re: [RFC] Btrfs device and pool management (wip)

2015-11-30 Thread Qu Wenruo
Austin S Hemmelgarn wrote on 2015/11/30 09:51 -0500: On 2015-11-30 02:59, Anand Jain wrote: Data center systems are generally aligned with the RAS (Reliability, Availability and Serviceability) attributes. When it comes to Storage, RAS applies even more because its matter of trust. In this

[RFC] Btrfs device and pool management (wip)

2015-11-30 Thread Anand Jain
(fixed alignment) Data center systems are generally aligned with the RAS (Reliability, Availability and Serviceability) attributes. When it comes to Storage, RAS applies even more because its matter of trust. In this context, one of the primary area that a typical volume manager

Re: [RFC] Btrfs device and pool management (wip)

2015-11-30 Thread Qu Wenruo
On 11/30/2015 03:59 PM, Anand Jain wrote: (fixed alignment) Data center systems are generally aligned with the RAS (Reliability, Availability and Serviceability) attributes. When it comes to Storage, RAS applies even more because its matter of trust. In this context, one of

Re: [RFC] Btrfs device and pool management (wip)

2015-11-30 Thread Austin S Hemmelgarn
On 2015-11-30 02:59, Anand Jain wrote: Data center systems are generally aligned with the RAS (Reliability, Availability and Serviceability) attributes. When it comes to Storage, RAS applies even more because its matter of trust. In this context, one of the primary area that a typical volume

[RFC] Btrfs device and pool management (wip)

2015-11-29 Thread Anand Jain
Data center systems are generally aligned with the RAS (Reliability, Availability and Serviceability) attributes. When it comes to Storage, RAS applies even more because its matter of trust. In this context, one of the primary area that a typical volume manager should be well