Is there a good way to debug a hanging NFS mount?

I would probalby start with do you have basic ip connectivity.

Yeah I have that. It's an etherstub and worked. But I noticed something in the /var/adm/messages svcprop: Pattern 'svc:/network/ipfilter:default' doesn't match any entities

So it looks like the Manifest script requires ipfilter but it's not installed

yep that was it... looks like a bug

Historical question: When did pkill(1) & pgrep(1) land in Solaris? It was *after* I got there, so 2.6 or later, but I can't remember exactly when.

7

danmcd: as far as I know, y'all are the only real NATIVE_ADJUNCT users, right?

richlowe: AFAICT yes.

danmcd: I'm going to have something for you to test that probably will involve selectively breaking that a bit.

(that is, a good test would include it not working, too)

Understood I think. I won't need to alter smartos-live, which SETS it:

configure:NATIVE_ADJUNCT="/opt/local"; export NATIVE_ADJUNCT

(Unless you want me to for breakage! :) )

I'm fixing some more of our native build stuff, and we'd want to check that your NATIVE_ADJUNCT stuff still triggers deflib if someone misses, which I think it's supposed to.

(but I must admit I'm not _sure_ it's supposed to)

So whatever you need I'll merge into illumos-joyent and see what happens.

@richlowe these bits ? code.illumos.org/c/illumos-gate/+/2762

→ CODE REVIEW 2762: 15552 NATIVE_LIBS needs a more truthful name (NEW) | illumos.org/issues/15552

danmcd: there's other bits coming to go with those

Ahhh.

Okay. ('cause if that was all I'd kick off the build now.)

unfortunately those other bits are confusing me greatly at the moment.

Understood.

[illumos-gate] 14421 use GCC 10 as default primary compiler -- Andy Fiddaman <illumos⊙fn>

hurray!

Please review the heads up message associated with this: illumos.topicbox.com/groups/develop…2/heads-up-moving-to-gcc-10-primary

if your build is insufficiently clobber-y, and your distribution has made certain choices, you can also expect to see a wodge of build noise as runpaths change.

Good point, and thanks rmustacc for your patience and for shepherding this in.

Eh, I did the easy stuff. It's really others efforts that are important here.