similar to: [PATCH v3 REPOST 0/5] threads: Add support for thread-safe handle.

Previously posted in 2015: v1: https://www.redhat.com/archives/libguestfs/2015-June/msg00048.html v2: https://www.redhat.com/archives/libguestfs/2015-June/msg00118.html I have rebased and tidied up the patches, fixing a few spelling mistakes, but they are broadly the same as before. I also ran all the tests, which pass. As with the previous versions, this makes a change to the API, where you

Previous discussion here: https://www.redhat.com/archives/libguestfs/2015-June/thread.html#00048 v2: - Use a cleanup handler to release the lock. - Rebase to upstream. Note I have not fixed the problem(s) with error handling (patch 3).

This patch isn't ready to go upstream. In fact, I think we might do a quick 1.30 release soon, and save this patch, and also the extensive changes proposed for the test suite[1], to after 1.30. Currently it is not safe to use the same handle from multiple threads, unless you implement your own mutexes. See: http://libguestfs.org/guestfs.3.html#multiple-handles-and-multiple-threads These

We permit the following constructs in libguestfs code: if (guestfs_some_call (g) == -1) { fprintf (stderr, "failed: error is %s\n", guestfs_last_error (g)); } and: guestfs_push_error_handler (g, NULL, NULL); guestfs_some_call (g); guestfs_pop_error_handler (g); Neither of these would be safe if we allowed the handle to be used from threads concurrently, since the error

We permit the following constructs in libguestfs code: if (guestfs_some_call (g) == -1) { fprintf (stderr, "failed: error is %s\n", guestfs_last_error (g)); } and: guestfs_push_error_handler (g, NULL, NULL); guestfs_some_call (g); guestfs_pop_error_handler (g); Neither of these would be safe if we allowed the handle to be used from threads concurrently, since the error

From: "Richard W.M. Jones" <rjones@redhat.com> With OCaml < 4.02 when using the alternate Bytes module, this module would be compiled twice during parallel builds, resulting in occasional corruption. The reason for this is that the ocamldep file mentions ‘bytes.cmo’ whereas the ‘$(OCAML_BYTES_COMPAT_ML)’ macro expands to ‘../../common/mlstdutils/bytes.ml’. Make doesn't

v3 -> v4: Use HAVE_OCAMLOPT in both places.

v1 was: https://www.redhat.com/archives/libguestfs/2017-July/msg00154.html There is no change in the first patch, but I added a second patch adding slow tests of opensuse guests (which pass, but require the first patch). Rich.

Removed extra $(OCAML_BYTES_COMPAT_CMO). Add if HAVE_OCAMLOPT around the call to ocamlopt. Tested with OCaml 4.01 on RHEL 7. Rich.

The C function mkdtemp(3) requires that the string ends with 6 'X' characters, so appending a non-empty suffix causes the function to raise EINVAL. Luckily we only ever called this function with the last parameter "". --- builder/builder.ml | 2 +- builder/sigchecker.ml | 2 +- common/mlutils/unix_utils.ml | 4 ++--

In the appliance used to build the packages for openSUSE, os-release is super minimal and only had ID_LIKE=suse. The code setting the DISTRO variable only searches for ID variable so far, resulting in invalid packagelist on openSUSE. This fix reads ID_LIKE as a fallback if ID contains nothing. --- m4/guestfs_appliance.m4 | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git

--- builder/sources.ml | 2 -- common/mlstdutils/Makefile.am | 4 ++- common/mlstdutils/stringSet.ml | 19 ++++++++++++ common/mlstdutils/stringSet.mli | 46 ++++++++++++++++++++++++++++ dib/elements.ml | 2 -- sysprep/sysprep_operation_ca_certificates.ml | 1 -

This is a simpler patch that removes GUESTFSD_EXT_CMD completely.

See: https://bugzilla.redhat.com/show_bug.cgi?id=1469655 https://github.com/systemd/systemd/issues/6334 https://lists.freedesktop.org/archives/systemd-devel/2017-July/039325.html

v1: https://www.redhat.com/archives/libguestfs/2017-July/msg00127.html v2: - Remove /etc/systemd/system/default.target.wants/guestfs-firstboot.service link as well. - Use multi-user.target for virt-p2v as well. Rich.

Untested - still working on reproducing this bug. Rich.

Even if ocamlopt is available, always build a bytecode version of ‘common/mlstdutils’. The generator is always built as bytecode. The generator depends on ‘../common/mlstdutils/guestfs_config.cmo’ and ‘../common/mlstdutils/std_utils.cmo’, but if these have not been built already then the generator races to build the .cmi files. Since the generator doesn't have the dependencies covering for

oraclelinux is the same as redhat/centos/scientificlinux, so add it where it is missing. This fixes amongst other things, running sysprep on an Oracle Linux image where it would previously fail operations like setting the hostname, saying that it was not supported on that distro. --- customize/firstboot.ml | 2 +- customize/hostname.ml | 4 ++--

In SUSE guests, handle the case where Bootloader::Tools::GetDefaultSection () returns undef. Previously this would return an empty string and cause a bogus error in subsequent code: virt-v2v: error: libguestfs error: statns: statns_stub: path must start with a / character --- v2v/linux_bootloaders.ml | 17 ++++++++++++++--- 1 file changed, 14 insertions(+), 3 deletions(-) diff --git

Even if ocamlopt is available, always build a bytecode version of ‘common/mlstdutils’. Furthermore, because this library is pure OCaml, we should not be using ‘ocamlmklib’. We should use ‘ocaml{c,opt} -a’ instead. This doesn't make any difference for native code, but for bytecode it was building a broken library. The original reason for making this change is because the generator is always