search for: pid1

...sock --filter=blocksize \ > +start_nbdkit -P blocksize2.pid -U blocksize2.sock --filter=blocksize \ > --filter=log file logfile=blocksize2.log blocksize2.img \ > minblock=1024 maxdata=512k maxlen=1M If the first process starts but the second fails, then you have not set pid1=, and leak the first process. This is a regression from the old code, which managed to capture pid1 before triggering the trap to cleanup(), and thus killed the successful nbdkit. Several tests are impacted. Less importantly, the old code performed startup in parallel (kick off both nbdkit p...

...tls.sh | 2 ++ tests/test-zero.sh | 13 ++++++++++++- 13 files changed, 36 insertions(+), 1 deletion(-) diff --git a/tests/test-blocksize.sh b/tests/test-blocksize.sh index 9cadf30..e008f5e 100755 --- a/tests/test-blocksize.sh +++ b/tests/test-blocksize.sh @@ -55,6 +55,8 @@ pid1= pid2= cleanup () { status=$? + trap '' INT QUIT TERM EXIT ERR + echo $0: cleanup: exit code $status test "$pid1" && kill $pid1 test "$pid2" && kill $pid2 diff --git a/tests/test-cache.sh b/tests/test-cache.sh index c46e7be..da1f6...

...> >+start_nbdkit -P blocksize2.pid -U blocksize2.sock --filter=blocksize \ > > --filter=log file logfile=blocksize2.log blocksize2.img \ > > minblock=1024 maxdata=512k maxlen=1M > > If the first process starts but the second fails, then you have not > set pid1=, and leak the first process. This is a regression from > the old code, which managed to capture pid1 before triggering the > trap to cleanup(), and thus killed the successful nbdkit. Several > tests are impacted. So one other idea I had is to have start_nbdkit create a trap function to...

...#!/usr/bin/env bash # nbdkit -# Copyright (C) 2019 Red Hat Inc. +# Copyright (C) 2019-2020 Red Hat Inc. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are @@ -64,14 +64,14 @@ files="$sock1 $sock2 $pid1 $pid2 nbd-tls-psk.out" rm -f $files cleanup_fn rm -f $files +# Run nbd plugin as intermediary; also test our retry code +start_nbdkit -P "$pid2" -U "$sock2" --tls=off nbd retry=5 \ + tls=require tls-psk=keys.psk tls-username=qemu socket="$sock1" + # Run enc...

...# test) if we ever changed that definition. + +source ./functions.sh +set -e +set -x + +requires jq --version +requires qemu-img --version +requires qemu-img map --help + +out="test-nbd-extents.out" +expected="test-nbd-extents.expected" +socket="test-nbd-extents.sock" +pid1="test-nbd-extents.pid1" +pid2="test-nbd-extents.pid2" +pid3="test-nbd-extents.pid3" +pid4="test-nbd-extents.pid4" +pid5="test-nbd-extents.pid5" +files="$out $expected $socket $pid1 $pid2 $pid3 $pid4 $pid5" +rm -f $files +cleanup_fn rm -f $...

...100755 --- a/tests/test-blocksize.sh +++ b/tests/test-blocksize.sh @@ -31,6 +31,7 @@ # OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF # SUCH DAMAGE. +source ./functions.sh set -e files="blocksize1.img blocksize1.log blocksize1.sock blocksize1.pid @@ -54,10 +55,6 @@ pid1= pid2= # Kill any nbdkit processes on exit. cleanup () { - status=$? - trap '' INT QUIT TERM EXIT ERR - echo $0: cleanup: exit code $status - test "$pid1" && kill $pid1 test "$pid2" && kill $pid2 # For easier debugging, dump the...

...bdinfo --version + +# Does the nbd plugin support dynamic lists? +if ! nbdkit --dump-plugin nbd | grep -sq libnbd_dynamic_list=1; then + echo "$0: nbd plugin built without dynamic export list support" + exit 77 +fi + +base=test-nbd-dynamic-list +sock1=`mktemp -u` +sock2=`mktemp -u` +pid1="$base.pid1" +pid2="$base.pid2" +files="$sock1 $sock2 $pid1 $pid2 $base.list $base.out1 $base.out2" +rm -f $files +cleanup_fn rm -f $files + +fail=0 + +# Start a long-running server with .list_exports and .default_export +# set to varying contents +start_nbdkit -P $pid...

...f with the plugin diff --git a/tests/test-blocksize.sh b/tests/test-blocksize.sh index 21b55b9..aa3cc13 100755 --- a/tests/test-blocksize.sh +++ b/tests/test-blocksize.sh @@ -50,14 +50,10 @@ if ! qemu-io -f raw -c 'r 0 1' -c 'w -z 1000 2000' \ fi truncate -s 10M blocksize2.img -pid1= pid2= - -# Kill any nbdkit processes on exit. +# For easier debugging, dump the final log files before removing them +# on exit. cleanup () { - test "$pid1" && kill $pid1 - test "$pid2" && kill $pid2 - # For easier debugging, dump the final log files...

v1 was here: https://www.redhat.com/archives/libguestfs/2018-September/msg00057.html v2: - Fix tab vs spaces in configure.ac. - To generate list of plugins, use printf instead of xargs. - Use 'source ./functions.sh' instead of 'source functions'. - functions.sh: Consistent quoting in foreach_plugin function. - functions.sh: Change the contract of start_nbdkit so it

Another series on top of my exportname filter, marking off another todo bullet point. With this, you can now use the NBD plugin as a transparent passthrough of all export names served by the remote server in both directions (list advertisement server to client, and export name from client to server). Eric Blake (3): nbd: Implement .default_export, .export_description nbd: Add

...like partial-dir or delay-updates. However I've found that a file was left after interruption, with a zero size. Would it be possible that rsync doesn't delete the partial file if it's not yet written to it? This is 3.0.6 on linux. Both processes were killed with the same command, kill pid1 pid2.

...POSSIBILITY OF +# SUCH DAMAGE. + +source ./functions.sh +set -xe + +requires nbdsh -c 'exit (not h.supports_uri ())' + +plugin=.libs/test-stdio-plugin.so +requires test -f $plugin + +sock1=`mktemp -u` +sock2=`mktemp -u` +files="test-stdio.in test-stdio.out test-stdio.err + test-stdio.pid1 test-stdio.pid2 $sock1 $sock2" +rm -f $files +cleanup_fn rm -f $files + +# Using a seekable file lets us prove that if the plugin consumes less +# than the full input, the next process sees the rest +cat >test-stdio.in <<EOF +string1 +string2 +EOF + +# .dump_plugin using stdout is nor...

...> -# Copyright (C) 2019 Red Hat Inc. > +# Copyright (C) 2019-2020 Red Hat Inc. > # > # Redistribution and use in source and binary forms, with or without > # modification, are permitted provided that the following conditions are > @@ -64,14 +64,14 @@ files="$sock1 $sock2 $pid1 $pid2 nbd-tls-psk.out" > rm -f $files > cleanup_fn rm -f $files > > +# Run nbd plugin as intermediary; also test our retry code > +start_nbdkit -P "$pid2" -U "$sock2" --tls=off nbd retry=5 \ > + tls=require tls-psk=keys.psk tls-username=qemu socket=&...

> > NB user namespaces are a fairly new piece of functionality that > is still somewhat rough around the edges. You usuaully want to > have the most recent kernel.org stable kernel available when > using this. > Yes, I confirm it is a little bit touchy for me. gabx@hortensia ➤➤ ~ % uname -a Linux hortensia 3.12.9-2-ARCH #1 SMP PREEMPT Fri Jan 31 10:22:54 CET 2014 x86_64

Combine much common code into tests/functions.sh. Patch 1: Preparation for patch 3. Patch 2: Fix a long-standing bug in how man pages links are generated. Patch 3: Common code for iterating a test function over every plugin. Patch 4: Common code for starting nbdkit in a test and waiting for the PID file to appear. This is the largest and most complex of the patches but is

...g file logfile=blocksize2.log blocksize2.img \ minblock=1024 maxdata=512k maxlen=1M - -# We may have to wait a short time for the pid files to appear. -for i in `seq 1 10`; do - if test -f blocksize1.pid && test -f blocksize2.pid; then - break - fi - sleep 1 -done - -pid1="$(cat blocksize1.pid)" || : -pid2="$(cat blocksize2.pid)" || : - -if ! test -f blocksize1.pid || ! test -f blocksize2.pid; then - echo "$0: PID files were not created" - exit 1 -fi +pid1="$(cat blocksize1.pid)" +pid2="$(cat blocksize2.pid)"...

...0M blocksize1.img +if ! $QEMU_IO -f raw -c 'r 0 1' -c 'w -z 1000 2000' \ + -c 'w -P 0 1M 2M' -c 'discard 3M 4M' blocksize1.img; then + echo "$0: missing or broken qemu-io" + rm blocksize1.img + exit 77 +fi +truncate --size 10M blocksize2.img + +pid1= pid2= + +# Kill any nbdkit processes on exit. +cleanup () +{ + status=$? + + test "$pid1" && kill $pid1 + test "$pid2" && kill $pid2 + # For easier debugging, dump the final log files before removing them. + echo "Log 1 file contents:" +...

...0M blocksize1.img +if ! $QEMU_IO -f raw -c 'r 0 1' -c 'w -z 1000 2000' \ + -c 'w -P 0 1M 2M' -c 'discard 3M 4M' blocksize1.img; then + echo "$0: missing or broken qemu-io" + rm blocksize1.img + exit 77 +fi +truncate --size 10M blocksize2.img + +pid1= pid2= + +# Kill any nbdkit processes on exit. +cleanup () +{ + status=$? + + test "$pid1" && kill $pid1 + test "$pid2" && kill $pid2 + # For easier debugging, dump the final log files before removing them. + echo "Log 1 file contents:" +...

...the basic .xml file: <domain type='lxc'> <name>dahlia</name> <memory>409600</memory> <os> <type arch='x86_64'>exec</type> <init>/bin/init</init> * systemd is default and PID1 is init on my Arch box <vcpu>4</vcpu> </os> <devices> <emulator>/usr/lib/libvirt/libvirt_lxc</emulator> * it is the path on Arch <console type='pty'/> </devices> </domain> Unfortunately : gabx@horte...

...dahlia</name> > <memory>409600</memory> > <os> > <type arch='x86_64'>exec</type> NB, you want 'exe' not 'exec' > <init>/bin/init</init> * systemd is > default and PID1 is init on my Arch box > <vcpu>4</vcpu> > </os> > <devices> > <emulator>/usr/lib/libvirt/libvirt_lxc</emulator> * it is the > path on Arch > <console type='pty'/> > </devices> > <...