01:02:30 aside from running iperf back and forth, anyone think of any 'fun' things to try to really kick the tires for a nic driver? 02:20:03 multiple parallel iperf runs while doing multiple rsyncs (mix of small and large files)? Maybe toss some NFS traffic in for variety. 02:28:46 jbk: I have recently discovered that rge can stall out and not recover in a way that I've only noticed when doing line rate IP forwarding that comes in the interface and goes back out the interface 02:28:48 So 02:29:00 Maybe try to use the thing as a router 02:29:42 iperf does not seem to excite the same issue, I suspect being an in-kernel hairpin like that just makes for tighter timing etc 03:26:00 ..... and after a while it stops passing unicast traffic (but broadcast traffice seems to get through) 03:26:06 i f$#@$@#$ hate this nic 03:30:23 jbk: I feel your pain 03:30:47 it's just so over complicated 03:32:27 and there is no hell hot enough for whoever decided on the design for their DMA engine 04:17:57 oh f me... the replacement drive is SMR.... no wonder why it's taken > 24 hours to resilver a bit over 1T of data 05:39:43 I'm sorry 14:17:59 yay.. managed to make it overnight without panicing (and still passing traffic) 16:19:02 oof (re: the SMR drive). 16:54:20 yeah :( lesson learned 17:34:21 put another way: fuck u *shingles ur magnetic recording* 18:10:11 the problem is there's no widely used filesystems (AFAIK) that really match how you want to use SMR 18:11:11 which basically (IIUC) your drive is divided into a number of regions, and basically you want to treat each region as 'append only' 18:14:35 and 'reset' the region to start over from its start when you want to rewrite it 18:15:20 so... use it like a tape drive? 18:49:12 i suppose sort of 18:50:16 i mean, if they're going to be around for a while, it might be worth thinking about such a fs... 18:51:13 though IIUC, there are _some_ similarities that could apply to flash as well (just apparently most flash devices have hw in front of the cells that do a better job of managing it on behalf of whatever is using it) 22:00:57 jbk: I'd think that something like the BSD log-structured file system might work ok with SMR if you could size the log segments to fit the SMR geometry. 22:34:50 https://www.openwall.com/lists/oss-security/2026/08/16/5 - AI can certainly build mile high walls of text no human brain can ever penetrate... 23:12:01 Well, metaslabs are actually pretty close. 23:48:32 I think one of them is saying that ZFS user authorizations do whoami such that you can direct them via namespacing? 23:48:50 but I'm far from sure 23:56:31 rmustacc: IIUC though, metaslabs use the 'free' space for accounting (shrinking as the metaslab utilization grows)... which would be less than ideal 23:59:32 (I suppose another problem is if you want to use mirror or raidz, the zone sizes would need to line up for it to work well..)