13:06:40 <tozhu> hello all, I hit / is full for my smartos, how to find out the core dump? anyone could advice?
13:18:47 <jperkin> check /root, 9 times out of 10 it's because you put something there
13:18:58 <jperkin> core dumps are configured to go into a separate file system so it won't be those
13:19:49 <tozhu> there is nothing under /root dir
13:20:31 <tozhu> the /root is empty, nothing is there
13:21:55 <jperkin> ncdu is available in the tools repo, that can be useful to see exactly where disk usage is, using e.g. "ncdu -x /" so that it only shows usage in the ramdisk
13:22:06 <tozhu> is there any core dump that fill the ramdisk space?
13:22:15 <jperkin> 13:18 < jperkin> core dumps are configured to go into a separate file system so it won't be those
13:23:13 <jperkin> everything on SmartOS is configured so that writes do not go to the ramdisk, have a think about anything you have done in the GZ recently that may have written a file to somewhere inside it
13:23:44 <tozhu> okay, thank you jperkin, I’m going to check it with ncdu
14:14:11 <tsoome> you can doublecheck the output of coreadm 
14:20:25 <tozhu> tsoome: thank you for the advice
16:32:37 <DarkUranium> Neat, a lot of people.
16:33:12 <DarkUranium> I've been wanting to evaluate SmartOS, but I'm struggling a bit with the instructions about non-interactive install ... they're telling me to put a file in `private/answers.json` on the installation media, but I can't tell how to mount said media to be able to make that file.
16:33:18 <DarkUranium> Which filesystem is the USB image?
16:33:32 <DarkUranium> (or can I simply add a random partition and make a directory there or something?)
16:42:10 <jperkin> DarkUranium: on the pcfs partition on the USB stick, though honestly for something I'm going to install once and then leave for possibly decades it only takes a few minutes to follow the install instructions and then I'm sure I did everything right
16:42:40 <DarkUranium> It's more a practical issue of not always having a KVM interface. (mostly just an issue of my test computer[s], but)
16:43:07 <DarkUranium> Basically, I don't have a display of any kind on the PC ATM. I could hook it up, but.
16:43:12 <DarkUranium> (I'm testing this on a random home PC)
16:47:01 <DarkUranium> Though I guess I'll just do that, heh.
16:54:42 <DarkUranium> jperkin: Oddly enough, though, all the various tools seem to be struggling with the partition table, reporting it as corrupt. Only gdisk managed to figure out the layout so far.
16:55:10 <DarkUranium> (and even that is reporting the GPT portion as "damaged")
17:00:32 <jperkin> that's odd, I mean it's been >10 years since I installed mine but I'm pretty sure it just mounted automatically after writing on macOS
17:01:13 <jperkin> hopefully someone else has some more recent experience ;)
17:02:25 <DarkUranium> gdisk gives me a "Caution: Found protective or hybrid MBR and corrupt GPT. Using GPT, but disk verification and recovery are STRONGLY recommended."
17:02:32 <DarkUranium> And it was the only one that even managed to give me a partition table.
17:02:35 <DarkUranium> So something's definitely funky.
17:03:01 <DarkUranium> (I tried all sorts of tools, even Windows' diskpart and stuff)
17:03:31 <DarkUranium> Right now, I'm extracting it with dd so that I can mount the loader partition with `mount` xD
17:03:37 <DarkUranium> (FreeBSD's mount doesn't have an `-o offset=...`)
17:12:53 <DarkUranium> jperkin: Dunno who needs to know, but gdisk has identified 4 problems with the partition table. For one, it's not GPT-compliant.
17:13:20 <DarkUranium> http://vpaste.net/pHpZL
17:14:39 <DarkUranium> Using the tool's recovery options has exposed the partitions to FreeBSD, finally (so that I get /dev/da0p{N} instead of just /dev/da0)
17:16:01 <DarkUranium> I'm pointing it out because it smells like a bug to me.
17:30:23 <sommerfeld> It's entirely possible that the FreeBSD GPT code is pickier than ours.
17:37:33 <DarkUranium> Definitely. Though like I said, I also tried in Windows (I didn't have a Linux machine handy to test on)
17:37:40 <DarkUranium> Though TBF, Windows is .... Windows.
17:40:31 <DarkUranium> sommerfeld: I suspect the core issue might've been the failing CRC checks. But gdisk fixed everything at once by using the backup table, so I can't really isolate that one issue.
17:40:46 <DarkUranium> Nor do I care to, TBH. It works now ^^