similar to: [PATCH libguestfs] guestfish: detect a few more failed syscalls

From: "Richard W.M. Jones" <rjones at redhat.com> For a test case, see: https://bugzilla.redhat.com/show_bug.cgi?id=803533 In guestfish, we use the gnulib closeout module which sets up an atexit handler to detect if there were any errors on stdout when the program exits. This is so we can fail correctly in out of disk space cases like: guestfish [...] > output However the

Hi Rich, I spend several days on this issue but still cannot figure out the root cause. Sometimes, guestfs_lauch() gets stuck, sometimes, it is ok. I reproduced this issue using guestfish: -bash-4.2# guestfish Welcome to guestfish, the guest filesystem shell for editing virtual machine filesystems and disk images. Type: 'help' for help on commands 'man' to read the manual

Thanks Rich. I have used libguestfs for several month. It worked perfectly before, the issue appears recently. I am trying guestfs_set_backend (g, "direct"). Thanks, Allen 2016-08-29 23:31 GMT+08:00 Richard W.M. Jones <rjones@redhat.com>: > On Mon, Aug 29, 2016 at 11:19:04PM +0800, Baochuan Wu wrote: > > Thanks Rich for you quick reply. I enabled logs and the program

The warnings in daemon were aggravating and risky for development (too easy to miss new ones) so I spent some time last week and today working on removing them. The first patch gets us down to almost no warnings with the original -Wall setting. That was by far the hardest part. Once I'd done that, I enabled nearly all of gcc's warnings via gnulib's warnings and manywarnings modules

Here is a bunch of small patches to make fish/ build with most warnings enabled: [1/9] edit.c: avoid warning about signed/unsigned comparison [2/9] fish.c: avoid warnings [3/9] tilde.c: avoid a warning [4/9] fish.c: avoid "assignment discards qualifiers..." warning [5/9] fish.c: avoid signed/unsigned-comparison warning [6/9] fish.c: don't perform arithmetic on void*

These patches first make src/ warning free, and then turn on the strict warning options. 75 0001-build-suppress-an-ignored-write-return-value-warning.patch 38 0002-build-suppress-an-ignored-dup-return-value-warning.patch 27 0003-generator.ml-suppress-signed-unsigned-compare-warnin.patch 48 0004-build-don-t-perform-arithmetic-on-void-pointers.patch 30

Using the latest code, I was seeing a failure of the remote alloc part of the test-remote.sh test: i.e., this would fail: make check -C regressions TESTS=test-remote.sh Running valgrind on it, I got this: $ libtool --mode=execute valgrind ../fish/guestfish --remote alloc test.img 10M ==11953== Memcheck, a memory error detector. ==11953== Copyright (C) 2002-2008, and GNU GPL'd, by

There were a few "infelicities" that kept the "make distcheck" test from passing. These three patches combine to fix all of them, punting on only one by disabling the install-data-hook rule. >From 7504acedcb71bd80d99abe412e6669b267cade38 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering at redhat.com> Date: Fri, 21 Aug 2009 13:37:18 +0200 Subject: [PATCH 1/3]

>From 9e99a1c6cb655a56d7f09dabd10a77a3802bf96d Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering at redhat.com> Date: Mon, 17 Aug 2009 18:44:37 +0200 Subject: [PATCH libguestfs 1/2] avoid compiler warnings about unused vars in generated code * src/Makefile.am: Compile protocol.c into a convenience library, so it can have its own CFLAGS, and link that with the destination one. ---

This might not be the final solution, but it certainly seems to solve a deadlock for me that I could trigger by using 'nbdkit --filter=noparallel memory 512k' and calling nbd_aio_pread for a request larger than 256k (enough for the Linux kernel to block the server until libnbd read()s), immediately followed by nbd_aio_pwrite for a request larger than 256k (enough to block libnbd until the

Nearly any program that writes to standard output can benefit from this sort of fix. Without it, running e.g., ./guestfish --version > /dev/full would exit successfully, even though it got ENOSPC when writing to the full device. That means regular output redirected to a file on a full partition may also fail to be written, and the error ignored. Before: $ guestfish --version >

On v1, we discussed whether cmds_to_issue needed to be a list, since it never had more than one element. I played with the idea of making it a list, and allowing the client to queue up new commands regardless of whether the state machine is currently in READY. I also polished up the tmp demo into a bit more full-fledged example file, worth including since it also let me discover a hard-to-hit race

Let's check for a quite a few more errors. Among other things, this adds some coverage for a few things I've patched recently. --- And these enhancements set me up for my next fix: making NBD_CMD_DISC prevent future commands. tests/errors.c | 167 +++++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 154 insertions(+), 13 deletions(-) diff --git a/tests/errors.c

On Thu, Jun 27, 2019 at 10:18:30PM -0500, Eric Blake wrote: > + /* Queue up a write command so large that we block on POLLIN, then queue > + * multiple disconnects. XXX The last one should fail. > + */ > + if (nbd_aio_pwrite (nbd, buf, 2 * 1024 * 1024, 0, 0) == -1) { > + fprintf (stderr, "%s: %s\n", argv[0], nbd_get_error ()); > + exit (EXIT_FAILURE); > +

On Monday 27 January 2014 13:29:54 Richard W.M. Jones wrote: > +/* Execute a command, sending output to a file. */ > +static int > +exec_command (char **argv, const char *file, int stderr_to_file) > +{ > + pid_t pid; > + int status, fd; > + FILE *fp; > + char line[256]; > + > + pid = fork (); > + if (pid == -1) { > + perror ("fork"); > +

On 6/30/19 12:54 PM, Richard W.M. Jones wrote: > On Thu, Jun 27, 2019 at 10:18:30PM -0500, Eric Blake wrote: >> + /* Queue up a write command so large that we block on POLLIN, then queue >> + * multiple disconnects. XXX The last one should fail. >> + */ >> + if (nbd_aio_pwrite (nbd, buf, 2 * 1024 * 1024, 0, 0) == -1) { >> + fprintf (stderr, "%s:

On Wed, Sep 11, 2019 at 08:48:28AM -0500, Eric Blake wrote: > On 9/11/19 5:40 AM, Richard W.M. Jones wrote: > > Very much a work in progress as there are still many tests using > > qemu-io which are candidates for conversion. > > > > You'll notice at the end of test-full.sh that the new test has some > > duplicated code which looks as if it ought to be

On Wed, Jun 19, 2019 at 09:11:52PM -0500, Eric Blake wrote: > When we are blocked waiting for POLLOUT during a request, and happen > to receive notice of POLLIN instead, we know that the work done in > response to POLLIN will be non-blocking (it returns to %.READY as soon > as it would block, which in turn jumps right back into ISSUE_COMMAND > because we have a pending request not

Very much a work in progress as there are still many tests using qemu-io which are candidates for conversion. You'll notice at the end of test-full.sh that the new test has some duplicated code which looks as if it ought to be refactored into a Python function. When I tried to do that, I got loads of strange Python problems which may indicate bugs in nbdsh itself or problems with my

In order to handle reading an in-flight response while in the middle of sending a second command, we'll need a way to jump back into the middle of a command being sent. This is easier if the state that sets wbuf is distinct from the state that reads into wbuf, and if we don't move the command to the in-flight queue until after the writes finish. --- generator/generator | 14