It's not full at the moment. Unless 79G is not enough for the next build.
slave167-c1:~$ df -h 
Filesystem      Size  Used Avail Use% Mounted on
udev            103G   12K  103G   1% /dev
tmpfs            21G  9.3M   21G   1% /run
/dev/sda1       493G  394G   79G  84% /
none            4.0K     0  4.0K   0% /sys/fs/cgroup
none            5.0M     0  5.0M   0% /run/lock
none            103G   40K  103G   1% /run/shm
none            100M     0  100M   0% /run/user

As for ssh-ing to slave167-c1, do you have the latest ssh config file and are you member of chrome-build-access?

I wouldn't be surprised if the build requires more than 79G. It's a ThinLTO build, and I think it builds all the binaries.

Is it possible to up the disk size of the VM?

hans, you should have access (I think), check the instructions here: https://chrome-internal.googlesource.com/infra/infra_internal/+/master/doc/ssh.md

If the machine is currently using up 394G of space, there's a bunch of stuff that can/should be deleted. We don't need more than 500GB per builder.

Hans, can you take a look first and then draft one of us if you get stuck?

(I was able to ssh now; it turns out I didn't know to use the .c.chromecompute domain)

There are a bunch of .tmp files lying around, which probably contributes to the problem. IIRC there's been changes to lld in that department lately:

$ ls -lh | grep .tmp | grep 'G '
-rwxr-xr-x  1 chrome-bot chrome-bot  1.9G Oct 31 06:10 blink_text_codec_UTF_8_fuzzer.tmp8464f3e
-rwxr-xr-x  1 chrome-bot chrome-bot  1.7G Nov  9 00:17 chrome_app_unittests.tmp6eff0ae
-rwxr-xr-x  1 chrome-bot chrome-bot  1.8G Nov 13 05:31 content_perftests.tmp89274e9
-rwxr-xr-x  1 chrome-bot chrome-bot  1.1G Nov 10 04:03 content_shell.tmp6e35402
-rwxr-xr-x  1 chrome-bot chrome-bot  1.9G Nov  3 14:54 extensions_browsertests.tmpb20dcd7
-rwxr-xr-x  1 chrome-bot chrome-bot  3.4G Nov  8 21:18 interactive_ui_tests.tmpf427f3e
-rwxr-xr-x  1 chrome-bot chrome-bot  2.6G Nov 11 16:09 metrics_unittests.tmp7f9101a
-rwxr-xr-x  1 chrome-bot chrome-bot  2.1G Oct 28 06:40 origin_trial_token_fuzzer.tmp7bd3742
-rwxr-xr-x  1 chrome-bot chrome-bot  2.2G Nov 13 05:31 renderer_fuzzer.tmpbc1933a
-rwxr-xr-x  1 chrome-bot chrome-bot  1.1G Nov 11 12:49 renderer_fuzzer.tmpd0e32e8
-rwxr-xr-x  1 chrome-bot chrome-bot  2.2G Nov 13 05:31 renderer_proto_tree_fuzzer.tmpbc3791f
-rwxr-xr-x  1 chrome-bot chrome-bot  1.2G Nov 12 13:21 views_examples_with_content_exe.tmp77670f3

The thin-lto cache takes up some space too:

$ du -sh thinlto-cache/
17G	thinlto-cache/

There are a lot of stale files in there:

$ ls -lrt | grep -v 'Nov 16' | wc -l
263736

I've nuked the out/Release dir on the slave. Let's see if that makes things better.

Hopefully LLVM's recent temp file handling changes will make the problem go away.

Looks like it's doing better now, except it's failing to upload size data.

dpranke: Can you help get 104.154.8.152 whitelisted?

> dpranke: Can you help get 104.154.8.152 whitelisted?

done.

Thanks!
Green build: https://build.chromium.org/p/chromium.clang/builders/ToTLinuxThinLTO/builds/763