similar to: are partial files always deleted on interruption?

We're still seeing sporadic failures of 'nbdkit nbd tls=', and I'm still trying to come up with a root cause fix (it may involve smarter use of gnutls_bye() in libnbd). In the meantime, here's what we know: when the hang/failure happens, the 'nbdkit nbd tls=' client process is stuck in a poll() waiting to see EOF from the server, while the 'nbdkit example1'

Borrows heavily from the existing test-truncate-extents.sh (basically, adding another layer of nbdkit to prove that the nbd plugin doesn't change anything from what the truncate filter is already tested to do). Signed-off-by: Eric Blake <eblake@redhat.com> --- tests/Makefile.am | 3 + tests/test-nbd-extents.sh | 113 ++++++++++++++++++++++++++++++++++++++ 2 files changed, 116

In these test functions, when a test fails, the bash ‘trap’ command causes the cleanup function to be called. However bash does not annul or cancel the traps when cleanup is called, so at the end of the cleanup function when ‘exit’ is called, cleanup is called recursively. Avoid this by cancelling the traps at the top of the cleanup function. Also an extra debugging message is emitted here

With new-enough libnbd and our new dynamic-export mode, we can grab the export list from the server for replay to the client. Signed-off-by: Eric Blake <eblake@redhat.com> --- plugins/nbd/nbdkit-nbd-plugin.pod | 7 ++ tests/Makefile.am | 2 + plugins/nbd/nbd.c | 53 ++++++++++ tests/test-nbd-dynamic-content.sh | 2 +- tests/test-nbd-dynamic-list.sh

Thanks: Eric Blake for the suggestion here: https://www.redhat.com/archives/libguestfs/2018-September/msg00069.html --- tests/functions.sh.in | 25 +++++++++++++++++++++++++ tests/test-blocksize.sh | 9 ++------- tests/test-cache.sh | 9 ++------- tests/test-cow.sh | 9 ++------- tests/test-data-7E.sh | 9

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

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

This assumes bashisms, but bash is required to run the tests. This is mostly refactoring. However the changes (simplifications) are quite substantial: - Since the new start_nbdkit helper function cleans up nbdkit on exit, most scripts no longer need to deal with the pid or kill the pid in the cleanup function. - As a result, cleanup functions are radically simpler, and often

On Thu, Mar 26, 2020 at 02:34:41PM -0500, Eric Blake wrote: > We're still seeing sporadic failures of 'nbdkit nbd tls=', and I'm > still trying to come up with a root cause fix (it may involve smarter > use of gnutls_bye() in libnbd). In the meantime, here's what we know: > when the hang/failure happens, the 'nbdkit nbd tls=' client process is > stuck in

Enhance the testsuite to ensure we don't regress with recent changes to stdin/out handling. This adds: - test-single-sh.sh: prove that 'nbdkit -s sh script' is viable - test-stdio.sh: create plugin that checks stdin/out match /dev/null, then run it with -s, --run, -f Signed-off-by: Eric Blake <eblake@redhat.com> --- tests/Makefile.am | 23 ++++++

The upstream NBD protocol recently clarified that servers can advertise block size limitations to clients that ask with NBD_OPT_GO (although we're still a ways off from implementing that in nbdkit); and that in the absence of that, then clients should agree on limits using out-of-band information or stick to sane defaults (everything 512-byte-aligned, no reads or writes larger than 32M). But

The upstream NBD protocol recently clarified that servers can advertise block size limitations to clients that ask with NBD_OPT_GO (although we're still a ways off from implementing that in nbdkit); and that in the absence of that, then clients should agree on limits using out-of-band information or stick to sane defaults (everything 512-byte-aligned, no reads or writes larger than 32M). But

This assumes bashisms, but bash is required to run the tests. This is mostly simple refactoring. Except for the test-memory*.sh tests where nbdkit used to run in the foreground, but that seems to be a consequence of some left over debugging. --- tests/functions.sh.in | 35 ++++++++++++++++++++ tests/test-blocksize.sh | 23 +++---------- tests/test-cache.sh

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

On 9/11/18 1:47 PM, Richard W.M. Jones wrote: > This assumes bashisms, but bash is required to run the tests. > > This is mostly simple refactoring. Except for the test-memory*.sh > tests where nbdkit used to run in the foreground, but that seems to be > a consequence of some left over debugging. > --- > +++ b/tests/functions.sh.in > @@ -32,6 +32,41 @@ > # OF THE

Originally, I thought that since the plugin always emulates .zero with a fallback to .pwrite, we didn't need to expose the backend's .can_zero to plugins. But there is another consideration, as shown at least in the nbd plugin: a plugin may implement a .zero callback that depends on support from a remote endpoint. If the remote endpoint doesn't support zeroes, then the plugin HAS to

Updated based on other changes that have happened in the meantime: - rely more on cleanup.h (throughout) - split structured read for easier review (patch 2 and 3 were combined in v1) - rely on nbdkit not leaking a server's partial answer (patch 3) - add tests (patch 5) - other bug fixes I found while testing it - drop EOVERFLOW patch for now; it will be separate once upstream NBD protocol

I thought I would forward this to see if anyone has any suggestions (other than, "You shouldn''t use this code in conjunction with fork"). Thanks, Dan ----- Forwarded Message ---- From: "Park, Chong SuhX" <chong.suhx.park at intel.com> To: Daniel Berger <djberg96 at yahoo.com> Sent: Monday, October 16, 2006 3:59:55 PM Subject: mmap 0.2.0 writing value

Dear all, I have a question on how to set up the starting model in POLYCLASS and make sure the terms in the starting model retained in the final POLYCLASS model. In the function POLYMARS, this can be done using the STARTMODEL option. See below for example, I started with model y= b0 + b1*X1 + b2*X2 + b3*X4 + b4*X5 + b5*X2*X5 + e > m00 <- matrix(c( 1, NA, 0, NA, 1, 2,

In v2: - use int instead of bool in the public header - split the tests from the code - don't overload test-layers; instead, add new tests - add a missing fflush exposed by the new tests - other minor cleanups Eric Blake (3): server: Add nbdkit_stdio_safe server: Sanitize stdin/out before running plugin code server: More tests of stdin/out handling docs/nbdkit-plugin.pod |