search for: set_export_name

...>meta_context_base_allocation); | assert (cmd == NBD_CMD_BLOCK_STATUS); | | blocks = extents_to_block_descriptors (extents, flags, count, offset, plus this sequence: $ patched/nbdkit memory 1M $ ./run nbdsh --opt-mode -u nbd://localhost nbd> h.set_request_meta_context(False) nbd> h.set_export_name('a') nbd> def c(arg): ... pass ... nbd> h.opt_set_meta_context_queries(['base:allocation'], c) 1 nbd> h.set_export_name('') nbd> h.opt_go() nbd> h.set_strict_mode(0) nbd> h.block_status(1024*1024, 0, c) nbdsh: generator/states-reply-chunk.c:425: enter_STA...

This replaces 13/13 of my v2 series; and now that it has pretty good testsuite coverage and demonstrable performance improvement to nbdinfo, I'm going ahead and pushing this now. We may still want to add further nbd_opt_* commands for other fine-grained tuning of negotiation, but for now, I think things have stabilized on this end, and I can return to polishing .list_exports on the nbdkit

...o ;; + open) echo "$3" ;; + export_description) echo "$2 world" ;; + get_size) echo 0 ;; + *) exit 2 ;; +esac +EOF + +# Without client request, nothing is advertised +nbdsh -c ' +import os + +h.set_opt_mode(True) +h.connect_unix(os.environ["sock"]) + +h.set_export_name("a") +h.opt_info() +try: + h.get_canonical_export_name() + assert False +except nbd.Error as ex: + pass + +h.set_export_name("") +h.opt_info() +try: + h.get_canonical_export_name() + assert False +except nbd.Error as ex: + pass + +h.opt_go() +try: + h.get_canonical_export...

...t first_version = [ > + "set_debug", (1, 0); > + "get_debug", (1, 0); > + "set_debug_callback", (1, 0); > + "clear_debug_callback", (1, 0); > + "set_handle_name", (1, 0); > + "get_handle_name", (1, 0); > + "set_export_name", (1, 0); > + "get_export_name", (1, 0); > + "set_tls", (1, 0); > + "get_tls", (1, 0); > + "set_tls_certificates", (1, 0); > + (*"get_tls_certificates", (1, 0);*) If we ever do implement this (and any other commented out fo...

...Quit the shell help (nbd) # Display documentation nbd> h.set_opt_mode(True) nbd> h.connect_command(["nbdkit","-s","--exit-","eval", "open=echo ENOENT >&2; exit 1"]) nbd> h.aio_is_negotiating() True nbd> h.set_export_name('a') nbd> h.opt_go() nbdkit: eval: error: /tmp/nbdkitNyQD8J/open: ENOENT nbd> h.set_export_name('b') nbd> h.opt_go() nbdkit: eval: error: /tmp/nbdkitNyQD8J/open: ENOENT nbd> h.set_opt_mode(True) Traceback (most recent call last): File "/usr/lib64/python3.8/code.py...

...ected tls mode' 2>&1; exit 1 + echo insecure; exit 0 + fi + echo $3 ;; + get_size) echo 6 ;; + pread) echo "$2" | dd skip=$4 count=$3 iflag=skip_bytes,count_bytes ;; + *) exit 2; +esac +EOF + +# Plaintext client sees only dummy volume +nbdsh -c ' +import os +h.set_export_name ("hello") +h.connect_unix (os.environ["sock"]) +assert h.get_size () == 512 +assert h.pread (5, 0) == b"dummy" +' + +# Encrypted client sees desired volumes +nbdsh -c ' +import os +h.set_export_name ("hello") +h.set_tls (nbd.TLS_REQUIRE) +h.set_tls_ps...

...uot; "test" "/" "//" " " \ > + "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx" > +do > + export e sock > + nbdsh -c ' > +import os > + > +e = os.environ["e"] > +h.set_export_name (e) > +h.connect_unix (os.environ["sock"]) Could be done with nbdsh --connect "nbd+unix://$e?socket=$sock". But what you did here works as well (--uri would work too once you could rely on nbdsh 1.2, but for now sticking to --connect is more portable) > + > +size = h...

...sh - <<'EOF' case "$1" in @@ -59,36 +58,34 @@ EOF nbdsh -c ' import os +def must_fail(f, *args, **kwds): + try: + f(*args, **kwds) + assert False + except nbd.Error: + pass + h.set_opt_mode(True) h.connect_unix(os.environ["sock"]) h.set_export_name("a") h.opt_info() -try: - h.get_canonical_export_name() - assert False -except nbd.Error as ex: - pass +must_fail(h.get_canonical_export_name) +must_fail(h.get_export_description) h.set_export_name("") h.opt_info() -try: - h.get_canonical_export_name() - assert False -e...

This doesn't include Eric's new APIs, but if you push those then I can rebase this one on top. Rich.

...t (cmd == NBD_CMD_BLOCK_STATUS); > | > | blocks = extents_to_block_descriptors (extents, flags, count, offset, > > plus this sequence: > > $ patched/nbdkit memory 1M > $ ./run nbdsh --opt-mode -u nbd://localhost > nbd> h.set_request_meta_context(False) > nbd> h.set_export_name('a') > nbd> def c(arg): > ... pass > ... > nbd> h.opt_set_meta_context_queries(['base:allocation'], c) > 1 > nbd> h.set_export_name('') > nbd> h.opt_go() > nbd> h.set_strict_mode(0) > nbd> h.block_status(1024*1024, 0, c) > nb...

..."xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx" +do + nbdsh -c ' +import os +import base64 + +e = os.environ["e"] +b = base64.b64encode(e.encode("utf-8")).decode("utf-8") +print ("e = %r, b = %r" % (e,b)) +h.set_export_name (b) +h.connect_unix (os.environ["sock"]) + +size = h.get_size () +assert size == len (e.encode("utf-8")) + +# Zero-sized reads are not defined in the NBD protocol. +if size > 0: + buf = h.pread (size, 0) + assert buf == e.encode("utf-8") +' +done + +# Test t...

...with a python connection to that socket: $ ~/qemu/qemu-nbd -r -k /tmp/nbdsock --object \ tls-creds-psk,id=tls0,endpoint=server,dir=/home/eblake/libnbd/tests \ --tls-creds tls0 -f raw -x / tmpfile $ ./run nbdsh nbd> h.set_tls_psk_file('tests/keys.psk') nbd> h.set_tls(2) nbd> h.set_export_name('/') nbd> h.connect_unix('/tmp/nbdsock') instead of getting a segfault. --- generator/generator | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/generator/generator b/generator/generator index c29460c..fa12232 100755 --- a/generator/generator ++...

...k "opt_list"; - Link "opt_info"]; + Link "opt_info"; Link "opt_list_meta_context"]; }; "get_opt_mode", { @@ -960,6 +965,56 @@ corresponding L<nbd_opt_go(3)> would succeed."; Link "set_export_name"]; }; + "opt_list_meta_context", { + default_call with + args = [ Closure context_closure ]; ret = RInt; + permitted_states = [ Negotiating ]; + shortdesc = "request the server to list available meta contexts"; + longdesc = "\ +Request that the serv...

...ames from the plugin. +for e in "" "test" "/" "//" " " \ + "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx" +do + export e sock + nbdsh -c ' +import os + +e = os.environ["e"] +h.set_export_name (e) +h.connect_unix (os.environ["sock"]) + +size = h.get_size () +print ("size=%r e=%r" % (size, e)) +assert size == len (e) + +# Zero-sized reads are not defined in the NBD protocol. +if size > 0: + buf = h.pread (size, 0) + print ("buf=%r e=%r" % (buf, e)) +...

..."base:allocation") > .set.export_name("/") > .set_debug(true) > .connect_uri("nbd://") > > And then you would be able to do, e.g.: > > h.pread(); > h.set_debug(false) > > But not: > > h.set_export_name("/exp"); > > as the object `h` would not even support that function. > > Anyway, even though it would be useful for libguestfs as well, thanks to > the way > these projects handle errors it is just something I consider > "nice-to-have". > > Have a nice...

Well, I'm not quite done (I still want to get nbdinfo to work on a single nbd connection for all cases when reading the heads of the file is not required), but I'm happy with patches 1-11, and 12-13 show where I'm headed for getting NBD_OPT_INFO to work. Posting now to see if some of the earlier patches are ready to commit while I continue working on the latter half. Eric Blake (13):

...st stable version was 1.2. + *) +let first_version = [ + "set_debug", (1, 0); + "get_debug", (1, 0); + "set_debug_callback", (1, 0); + "clear_debug_callback", (1, 0); + "set_handle_name", (1, 0); + "get_handle_name", (1, 0); + "set_export_name", (1, 0); + "get_export_name", (1, 0); + "set_tls", (1, 0); + "get_tls", (1, 0); + "set_tls_certificates", (1, 0); + (*"get_tls_certificates", (1, 0);*) + "set_tls_verify_peer", (1, 0); + "get_tls_verify_peer", (1, 0)...

...> + > +# Test that it fails if the caller passes in non-base64 data. The > +# server drops the connection in this case so it's not very graceful > +# but we should at least get an nbd.Error and not something else. > +nbdsh -c ' > +import os > +import sys > + > +h.set_export_name ("xyz") > +try: > + h.connect_unix (os.environ["sock"]) > + # This should not happen. > + sys.exit (1) > +except nbd.Error as ex: > + sys.exit (0) > +# This should not happen. > +sys.exit (1) The test looks good. (Yeah, figuring out how to m...

The previous incomplete patch was here: https://www.redhat.com/archives/libguestfs/2019-September/msg00049.html based on earlier discussion here: https://www.redhat.com/archives/libguestfs/2019-September/msg00047.html In v2: - The previous patch was incomplete. This version completes it by adding tests and extending nbdkit-sh-plugin. - nbdkit_export_name now returns NULL for error,

This is a revision of my .default_export work, plus new work on .export_descriptions and a new exportname filter. I think it is now ready to check in. Things I'd still like in 1.22: - the file plugin should implement .list_exports (patch already posted, but it needs rebasing on this series) - the ext2 filter should override .list_exports when in exportname mode - the nbd plugin should be