jbk: Right, I'm saying you could make a service that could do all kinds of diagnostics things like that if you were easily able to subscribe to SMF state changes

We have some nascent things in the Oxide control plane software that poke at bits of system state for collecting evidence about potential issues

I think generally it's going to be pretty hard to make a generic service that does what everyone wants, as a lot of it is going to come down to making policy decisions

But we could much more easily provide really good first class interfaces that people could consume in their appliance- or site-specific tools

yeah, just thinking out loud (online?)

We all do it once or twice.

Hey @jbk ==> illumos.org/issues/17863

→ BUG 17863: libsff_8472 test needs update to libsff_8472.out post-17827 (New)

I think you forgot to update the libsff_8472 expected-output.

hrm.. yeah.. let me see if I can get a fix up this afternoon.. meetings this morning

danmcd: I'm assuming I can use 17863? I can put it up on gerrit, but will need to go back over everything after lunch

Yeah, you should be good to use it jbk.

Dan didn't assign himself.

ok.. i'll take it then

17863 the binary is quite self-contained, and the diff pipeline I put up there is the perfect way to confirm/deny it.

SHould be just mods to the .out file.

i've put up a fix on gerrit, it looks reasonable, but I want to sit down after lunch just to go over the changes with the changes in the code to be sure before I call it good..

I'll hopefully stop getting bombarded with multiple things all at once so I can focus :) *fingers crossed*

Heh heh heh. I'm supplying you with a full release-week test run (SmartOS gcc14 in this case) where I can compare it against its prior release.

[illumos-gate] 17847 zfs: add ability to set user properties while changing encryption key -- Matthew Ahrens <matt⊙mo>

doh! SFF_8636_ENHANCED_OPTIONS_MASK should be 0x1f not 0x1e..

i guess that'll need a follow up change (but thankfully a small one)

@jbk to your 17863 changes?

yeah..

danmcd: mind if I squash that fix so it's one review (w/ the update fixes)?

that way it can be merged together

Does the second one have major impact? I didn't notice any in my original runs, just the libsff ones.

Ahhh I see. Yeah, squish 'em.

not really, it just fixes thigns up

2 files changed, 9 insertions(+), 7 deletions(-)

Would it be a bad idea to install OmniOS and then put a desktop environment on that?

i don't know if it'd be a bad idea, but yuo might have to do more work since it's not really geared towards that...

I'm just not really sure I want to try OpenIndiana instead.

The comparison table on illumos.org says it approximately compares to Ubuntu.

I'm deliberately not using Ubuntu.

msiism: well, the ammount of work to add a desktop to omnios is .... huge. this is exactly what openindiana does for you. I would not compare ubuntu with openindiana.

Okay. :)

just building all the desktop bits is a lot of annoying work, usually

I think I've done it two or three times, ever, and when it was easier.

I see.

anubis on code.illumos.org is not letting me past the "making sure you're not a bot" page.

anyone else seeing this?

Is it asking you to activate Javascript?

sommerfeld: it is working harder now, because the bots got around it and ruined everything

but you should get past it if you're human

jclulow: do we have logs?

the only change I saw was that 1) the service was up again, and 2) I actually had to be vetted by anubis now

so I got past it by deleting illumos.org cookies.

weird

jclulow: seems definitely something you'd want to know

perhaps I had something stale in cookies that confused things and sent it in a redirect-to-self loop.

and after signing in again it's still working so I'm unblocked.

richlowe: We do have logs, and they're entirely useless

But yeah, there is clearly something happening with the cookies. I am going to look at it again this afternoon

If you clear the cookies, for now, that seems to get past it.

As ever: we apologise for the inconvenience.