search for: cleanup_fn

...changed, 73 insertions(+), 157 deletions(-) diff --git a/tests/functions.sh.in b/tests/functions.sh.in index 62fdfe5..42e3925 100644 --- a/tests/functions.sh.in +++ b/tests/functions.sh.in @@ -32,6 +32,31 @@ # OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF # SUCH DAMAGE. +# cleanup_fn f [args] +# +# A generic trap handling function. This runs the function f + args +# when the script exits for any reason. +declare -a _cleanup_hook +cleanup_fn () +{ + _cleanup_hook[${#_cleanup_hook[@]}]="$@" +} + +_run_cleanup_hooks () +{ + status=$? + set +e + trap '...

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

...{1..10}; do + if test -f "$pidfile"; then + break + fi + sleep 1 + done + if ! test -f "$pidfile"; then + echo "$0: start_nbdkit: PID file $pidfile was not created" + exit 1 + fi + + # Kill nbdkit on exit. + cleanup_fn kill "$(cat $pidfile)" +} + # foreach_plugin f # # For each plugin that was built, call test function 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...

v1 was here: https://www.redhat.com/archives/libguestfs/2019-January/thread.html#00198 For v2 I changed most existing prerequisite tests to use the new mechanism. I only changed simple tests. There are a few more complex tests that don't fit the “requires model” and those are not changed. I normalized qemu-io/qemu-img testing to always use the --version flag, where previously we used a mix

...-git a/tests/test-data-7E.sh b/tests/test-data-7E.sh index e27984f..2e48e1b 100755 --- a/tests/test-data-7E.sh +++ b/tests/test-data-7E.sh @@ -38,16 +38,12 @@ source ./functions.sh set -e set -x +requires qemu-io --version + files="data-7E.out data-7E.pid data-7E.sock" rm -f $files cleanup_fn rm -f $files -# Test that qemu-io works -if ! qemu-io --help >/dev/null; then - echo "$0: missing or broken qemu-io" - exit 77 -fi - # Run nbdkit. start_nbdkit -P data-7E.pid -U data-7E.sock --export= \ --filter=partition \ diff --git a/tests/test-data-base64.sh b/tes...

On 9/13/18 11:09 AM, Richard W.M. Jones wrote: > Thanks: Eric Blake for the suggestion here: > https://www.redhat.com/archives/libguestfs/2018-September/msg00069.html > --- I'm glad the idea worked! > +# cleanup_fn f [args] > +# > +# A generic trap handling function. This runs the function f + args Not just functions, but any command name... > +# when the script exits for any reason. > +declare -a _cleanup_hook > +cleanup_fn () > +{ > + _cleanup_hook[${#_cleanup_hook[@]}]="$@&q...

...ot; > + echo "$0: test prerequisite is missing or not working" > + exit 77 > + } > +} Updated function looks good. > +++ b/tests/test-ip.sh > @@ -37,26 +37,17 @@ > source ./functions.sh > set -e > > -rm -f ip.pid ipv4.out ipv6.out > -cleanup_fn rm -f ip.pid ipv4.out ipv6.out > +requires ss --version > +requires ip -V > +requires qemu-img --version > > -# Don't fail if certain commands aren't available. > -if ! ss --version; then > - echo "$0: 'ss' command not available" > - exit 77...

...not supported"; then - echo "$0: IPv6 loopback is not available, skipping this test" - exit 77 + echo "$0: IPv6 loopback is not available, skipping this test" + exit 77 fi } @@ -139,7 +139,32 @@ start_nbdkit () fi # Kill nbdkit on exit. - cleanup_fn kill "$(cat "$pidfile")" + cleanup_fn kill_nbdkit "$(cat "$pidfile")" +} + +# kill_nbdkit pid +# +# End the nbkdit process that created pidfile. Exit this script with an +# error if nbdkit does not gracefully shutdown in a timely manner. +kill_nbdkit () +...

This rounds up the remaining bugs that I originally identified in: https://www.redhat.com/archives/libguestfs/2020-July/msg00153.html Eric Blake (4): api: Permit export list APIs when Connected info: Support --list with serializing servers info: Fix --json output when list size != 1 info: Permit --size --json generator/API.ml | 6 +++--- info/info-list-json.sh | 9 +++++++++

...fo/info-list.sh | 3 ++- info/nbdinfo.c | 21 ++++++++++++++++----- 3 files changed, 20 insertions(+), 7 deletions(-) diff --git a/info/info-list-json.sh b/info/info-list-json.sh index 3845875..af29064 100755 --- a/info/info-list-json.sh +++ b/info/info-list-json.sh @@ -33,7 +33,7 @@ cleanup_fn rm -f $img $out $pid $sock rm -f $img $out $pid $sock truncate -s 1M $img -qemu-nbd -t --socket=$sock --pid-file=$pid -x "hello" $img & +qemu-nbd -t --socket=$sock --pid-file=$pid -x "hello" -D "world" $img & cleanup_fn kill $! # Wait for qemu-nbd to star...

On Wed, Jan 23, 2019 at 10:02:18AM -0600, Eric Blake wrote: > > +++ b/tests/test-ip.sh > > @@ -37,26 +37,17 @@ > > source ./functions.sh > > set -e > > > > -rm -f ip.pid ipv4.out ipv6.out > > -cleanup_fn rm -f ip.pid ipv4.out ipv6.out > > +requires ss --version > > +requires ip -V > > +requires qemu-img --version > > > > -# Don't fail if certain commands aren't available. > > -if ! ss --version; then > > - echo "$0: 'ss' command no...

I only did a handful of the tests to demonstrate the point. If the patch approach is accepted then I'll do the remainder of the tests in v2. Rich.

...ests/test-memory-largest.sh index a6c269e..50708f9 100755 --- a/tests/test-memory-largest.sh +++ b/tests/test-memory-largest.sh @@ -37,16 +37,12 @@ source ./functions.sh set -e +requires qemu-io --help + files="memory-largest.out memory-largest.pid memory-largest.sock" rm -f $files cleanup_fn rm -f $files -# Test that qemu-io works -if ! qemu-io --help >/dev/null; then - echo "$0: missing or broken qemu-io" - exit 77 -fi - # Run nbdkit with memory plugin. # size = 2^63-1 start_nbdkit -P memory-largest.pid -U memory-largest.sock \ diff --git a/tests/test-partition...

Now that we have implemented systemd socket activation, we can use this to run qemu-nbd in tests. The first patch leaves some dead code around (the -DSERVE_OVER_TCP=1 path). It's possible we might want to use this to test against a putative future NBD server that only supports TCP, but on the other hand maybe we should just remove it. Tests & valgrind still pass for me. Rich.

...t -e +set -ex # Determine the absolute pathname of the execvpe helper binary. The "realpath" # utility is not in POSIX, but Linux, FreeBSD and OpenBSD all have it. @@ -155,7 +155,7 @@ success() # Create a temporary directory and change the working directory to it. tmpd=$(mktemp -d) -cleanup_fn rm -r -- "$tmpd" +#cleanup_fn rm -r -- "$tmpd" cd "$tmpd" # If the "file" parameter of execvpe() is an empty string, then we must fail -- $ podman build -f ci/containers/alpine-edge.Dockerfile -t libnbd-alpine-edge $ podman run -it --rm --userns=keep-id -...

...he absolute pathname of the execvpe helper binary. The "realpath" > # utility is not in POSIX, but Linux, FreeBSD and OpenBSD all have it. > @@ -155,7 +155,7 @@ success() > > # Create a temporary directory and change the working directory to it. > tmpd=$(mktemp -d) > -cleanup_fn rm -r -- "$tmpd" > +#cleanup_fn rm -r -- "$tmpd" > cd "$tmpd" > > # If the "file" parameter of execvpe() is an empty string, then we must fail -- > > $ podman build -f ci/containers/alpine-edge.Dockerfile -t libnbd-alpine-edge > $ podma...

...ecvpe helper binary. The "realpath" > > # utility is not in POSIX, but Linux, FreeBSD and OpenBSD all have it. > > @@ -155,7 +155,7 @@ success() > > > > # Create a temporary directory and change the working directory to it. > > tmpd=$(mktemp -d) > > -cleanup_fn rm -r -- "$tmpd" > > +#cleanup_fn rm -r -- "$tmpd" > > cd "$tmpd" > > > > # If the "file" parameter of execvpe() is an empty string, then we must fail -- > > > > $ podman build -f ci/containers/alpine-edge.Dockerfile -t l...

...eal.sh @@ -1,6 +1,6 @@ #!/usr/bin/env bash # nbdkit -# Copyright (C) 2018-2019 Red Hat Inc. +# Copyright (C) 2018-2020 Red Hat Inc. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions are @@ -51,15 +51,7 @@ cleanup_fn rm -f $files qemu-img create -f vmdk test-vddk-real.vmdk 100M -export old_ld_library_path="$LD_LIBRARY_PATH" -export LD_LIBRARY_PATH="$vddkdir/lib64:$LD_LIBRARY_PATH" - nbdkit -f -v -U - \ --filter=readahead \ vddk libdir="$vddkdir" file=test-vddk...

...nged, 204 insertions(+) > > +# Test the info plugin with mode=conntime. > + > +source ./functions.sh > +set -e > +set -x > + > +requires nbdsh --version > + > +sock=`mktemp -u` > +files="info-conntime.out info-conntime.pid $sock" > +rm -f $files > +cleanup_fn rm -f $files > + > +# Run nbdkit. > +start_nbdkit -P info-conntime.pid -U $sock info mode=conntime > + For conntime, is it worth testing 2 separate connections: Connection 1: connect, sleep 1, read Connection 2: connect, read where the delta read by connection 2 should be less than t...

We were previously dlopen-ing it in the load() method. This is very early and in particular means that the only possible way to configure where we find the library is through environment variables and not through config parameters. Also it's not necessary as we don't call any functions from the library (such as VixDiskLib_InitEx) until config_complete. This change is neutral refactoring