07:11:57 <tsoome> honest manual: BUGS This program does not work very well.
08:35:31 <tsoome> anyhow, I did a small test, picked MOS DVA, nuked with /dev/zero block pointed by dva[0], observed loader getting checksum error and picking dva[1] instead, but also zpool import poolname was all happy - the only bug was that self repair event is not recorded for it -- the block was repaired silently, but zpool status does not reflect the checksum error
19:46:30 <richlowe> tsoome: could we make loader say something if it has to read non-primary blocks?
19:46:58 <richlowe> or is that a thing were just supposed to silently succeed? I reckon I'd probably want to know
19:48:14 <tsoome> sure we can, currently is just walking through dva list and is only complaining when all reads have failed.
19:48:50 <tsoome> and yep, its really about noise... 
20:02:39 <richlowe> tsoome: yeah, I'm never good at balancing noise v. action
20:02:51 <richlowe> but I think anything weird about my storage I'd want a clue as early as possible, just not it spammed
20:03:09 <richlowe> (for eg, my windows experience when disks fail is most of the I/O that fails is the windows event logs saying the disk is failing...)
20:04:24 <tsoome> I actually do think it is good idea. 
21:19:38 <sommerfeld> richlowe: periodic scrubs should catch and repair that sort of thing.
21:20:26 <sommerfeld> (loader presumably doesn't attempt to repair the busted copy of the block..)
21:20:42 <tsoome> no, it does not.
21:24:46 <sommerfeld> maybe the way to go is for the loader to remember that this happened and somehow leave a note for the kernel.   "drop this into the FMA error log and let the ZFS diagnosis engine do something about it.."
21:25:27 <richlowe> (boot module of nvlist?)
21:28:49 <jbk> i do have a change to improve sd + zfs.. at least for supported SCSI disks (enterprise ones seem to) where a read fails due to a bad block will trigger remapping of the bad block and then a zfs self-heal
21:29:09 <jbk> (today, zfs self healing only works if the drive returns a _successful_ I/O with invalid data)
21:29:48 <jbk> but i've been working on a userland iscsi client (based off a rust crate) to simulate things for testing
21:32:58 <tsoome> sommerfeld nice one;) 
21:42:49 <sommerfeld> jbk: yeah, having simulated hardware that can fail on demand is crucial for things like that..
21:49:33 <jbk> just E810 is like a bottomless pit of code
21:50:34 <jbk> i thought i was almost done, and then 'no, to get RSS working so more than one RX ring is actually used, it requires thousands of lines of additional code'
21:55:01 <jbk> (20k lines to date)
22:04:20 <sommerfeld> ow.
22:31:57 <gitomat> [illumos-gate] Backout: 18149 ld(1) should support -zsysroot -- Richard Lowe <richlowe⊙rn>
22:49:25 <richlowe> I think a bare minimum of you were in a position to experience that, but if you were you have my apologies 
23:15:42 <sommerfeld> been in tree since August 27th?
23:24:59 <richlowe> I'm told that in practice it doesn't happen through the gcc frontends, you would have to have that commit and be using (or possibly building?) clang