Monday, August 27, 2018

Back from a nice vacation / conference trip.

A nice trip

Visited Boston for DevConf.us where I was giving three talks (one on vGPUs, one on make-it-quick and one on recorder). Took a little detour through Westford to see the Red Hat offices there and visit my managers and a few other people.

Spent some time with friends Tristan and Edith Mollet in Caldwell. Spent one day visiting Daveed and Karina Vandevoorde near Princeton. Also a quick day in New York, Ellen's Stardust restaurant, Time Square, Central Park, etc. Can't say I really love New York but it was nice for the kids to see that.

The circuit breakers had tripped a couple of days after we left due to heavy rain (this often happens when there are thunderstorms). Predictably, the fridge and freezer were full of really foul-smelling stuff. Spent a few hours cleaning up.

Another bad surprise was that my E-Mehari electric car failed to wake up from deep sleep. They must share that part of the code with the MacBook which also regularly fails to wake up these days. I had to go pick up a rental car this morning, since Christine leaves tomorrow so I need a car just in case.

SMB crash

Got repeated failures connecting to Turbo over SMB (e.g. for backup). Looking at /var/log/samba/log.smbd, I see repeated crashes that look like this:

  PANIC (pid 29881): could not store share mode entry: NT_STATUS_UNSUCCESSFUL
[2018/08/27 13:01:29.563422,  0] ../source3/lib/util.c:926(log_stack_trace)
  BACKTRACE: 29 stack frames:
   #0 /lib64/libsmbconf.so.0(log_stack_trace+0x23) [0x7f9b137e39e3]
   #1 /lib64/libsmbconf.so.0(smb_panic_s3+0x24) [0x7f9b137e3ac4]
   #2 /lib64/libsamba-util.so.0(smb_panic+0x33) [0x7f9b158f42d3]
   #3 /usr/lib64/samba/libsmbd-base-samba4.so(+0x1f119a) [0x7f9b1554819a]
   #4 /lib64/libtalloc.so.2(+0xaf06) [0x7f9b12602f06]
   #5 /lib64/libtalloc.so.2(+0xa820) [0x7f9b12602820]
   #6 /lib64/libtalloc.so.2(_talloc_free+0xd8) [0x7f9b125fd658]
   #7 /usr/lib64/samba/libsmbd-base-samba4.so(+0x17f2ee) [0x7f9b154d62ee]
   #8 /usr/lib64/samba/libsmbd-base-samba4.so(+0x181460) [0x7f9b154d8460]
   #9 /usr/lib64/samba/libsmbd-base-samba4.so(create_file_default+0x176) [0x7f9b154d9ca6]
   #10 /usr/lib64/samba/vfs/fruit.so(+0x8698) [0x7f9af40cb698]
   #11 /usr/lib64/samba/libsmbd-base-samba4.so(smbd_smb2_request_process_create+0xaab) [0x7f9b1550fa8b]
   #12 /usr/lib64/samba/libsmbd-base-samba4.so(smbd_smb2_request_dispatch+0x144e) [0x7f9b15506ebe]
   #13 /usr/lib64/samba/libsmbd-base-samba4.so(+0x1b161d) [0x7f9b1550861d]
   #14 /lib64/libtevent.so.0(+0xaa7a) [0x7f9b121e5a7a]
   #15 /lib64/libtevent.so.[2018/08/27 13:01:29.563851,  0] ../source3/lib/dumpcore.c:318(dump_core)
  coredump is handled by helper binary specified at /proc/sys/kernel/core_pattern[2018/08/27 13:01:29.770809,  0] ../source3/locking/share_mode_lock.c:439(share_mode_data_destructor)
  store returned NT_STATUS_UNSUCCESSFUL

Did not file a BZ yet, it looks like there is a DNF update.

DNF update does not go well

There are errors with the DNF update:

  Curl error (23): Failed writing received data to disk/application for http://ftp.lip6.fr/ftp/pub/linux/distributions/fedora/updates/28/Everything/x86_64/Packages/l/libreoffice-core-6.0.6.2-1.fc28.x86_64.rpm [Failed writing body (746 != 1448)]

and then the next attempt gives me:

[MIRROR] mozjs52-52.9.0-1.fc28.x86_64.rpm: Curl error (23): Failed writing received data to disk/application for http://distrib-coffee.ipsl.jussieu.fr/pub/linux/fedora/linux/updates/28/Everything/x86_64/Packages/m/mozjs52-52.9.0-1.fc28.x86_64.rpm [Failed writing body (927 != 1448)]

Not sure if the issue is remote or local disk. The message is not very clear. It says "error downloading package" but then "failed writing body".

Apparently, I still this problem where /data did not mount correctly and it filled my root device:

# df -h
Filesystem               Size  Used Avail Use% Mounted on
[...]
/dev/mapper/fedora-root   50G   49G     0 100% /
[...]

After removing close to 30G in /data/music (probably backup from io before /data/ was mounted), I get a more reasonable content for the root device:

/dev/mapper/fedora-root   50G   20G   27G  43% /

I need to find a way to stop io backups to fill up the root partition like this.

dnf udpate hangs

The next problem is that dnf update hangs mid-flight with:

  Upgrading        : flatpak-1.0.0-2.fc28.x86_64                       373/1950
  Running scriptlet: flatpak-1.0.0-2.fc28.x86_64                       373/1950

(Stuck there for over one hour)

On the console, I have a probably unrelated message:

[81228.843502] systemd[1]: /usr/lib/systemd/system/virtlockd-admin.socket:7: Unknown lvalue 'Server' in section 'Socket'

Filtering, filtering

Added a few dozen new rules to filter my incoming mail. Today, I spent way too long filtering low-priority email. Hopefully, tomorrow will be easier.