netcf-devel

netcf-devel@lists.fedorahosted.org

270 discussions

Re: RFC: deprecating/obsoleting netcf package and libvirt virInterface*() APIs
by Daniel P. Berrangé 03 Dec '20

03 Dec '20

On Wed, Dec 02, 2020 at 02:47:53PM -0500, Laine Stump wrote: > 1) Does this generally sound like a good direction? Or is there something > I'm ignoring that renders my points moot? > > 2) If we are going to do it, how should we proceed? > > We obviously can't simply *remove* the virInterface API from libvirt (since > that would destroy backward compatibility guarantees), but could immediately > begin logging some sort of "this API is deprecated" message when any of the > functions are called, and then in a later release change the APIs to return > an error (while simultaneously removing netcf from the build and dependency > lists). At the same time, we would need to decide if the "interface status" > functionality needs to be maintained within appropriate virInterface*() > APIs, reproduced in virNodeDeviceGetXMLDesc(), or just dropped altogether. We should *NOT* deprecate the public APIs. The design of them is perfectly valid and can accept any other implementation in future. A more viable impl of network APIs could be to back them by Networkmanager, or systemd network, or whatever is relevant to a particular platform If we want to drop netcf because the project is unmaintained that is fine, the APIs simply return an VIR_ERR_NO_SUPPORT code as with any other API that has no impl in a particular hypervisor. Regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|

1 0

[PATCH 0/3] A few fixes to placate Coverity
by Laine Stump 17 Sep '18

17 Sep '18

Admittedly patch 1 was a legitimate leak.... Laine Stump (3): Fix memory leak in aug_match_mac() remove unused variable in aug_match_mac() Make the empty body of for loops more obvious src/dutil_linux.c | 4 ++-- src/ncftool.c | 9 ++++++--- 2 files changed, 8 insertions(+), 5 deletions(-) -- 2.14.4

2 4

[PATCH] build: Silence gcc warnings
by Eric Blake 06 Sep '18

06 Sep '18

The first warning is obvious: xslt_ext.c: In function 'ipcalc_netmask': xslt_ext.c:72:53: warning: format '%d' expects argument of type 'int', but argument 5 has type 'long unsigned int' [-Wformat=] "ipcalc:netmask: prefix %d not in the range 1 to 32", prefix); ~^ ~~~~~~ %ld The second warning looks like it was intended to have a const pointer to const contents, where the second 'const' was misplaced instead of duplicated: ncftool.c:96:33: warning: duplicate 'const' declaration specifier [-Wduplicate-decl-specifier] static const struct command_def const *commands[]; ^~~~~ ncftool.c:676:33: warning: duplicate 'const' declaration specifier [-Wduplicate-decl-specifier] static const struct command_def const *commands[] = { ^~~~~ Signed-off-by: Eric Blake <eblake(a)redhat.com> --- src/ncftool.c | 4 ++-- src/xslt_ext.c | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/src/ncftool.c b/src/ncftool.c index f1b5642..382c91d 100644 --- a/src/ncftool.c +++ b/src/ncftool.c @@ -93,7 +93,7 @@ struct command_opt { /* Global variables */ -static const struct command_def const *commands[]; +static const struct command_def *const commands[]; struct netcf *ncf; static const char *const progname = "ncftool"; @@ -673,7 +673,7 @@ static int parseline(struct command *cmd, char *line) { return 0; } -static const struct command_def const *commands[] = { +static const struct command_def *const commands[] = { &cmd_list_def, &cmd_dump_xml_def, &cmd_define_def, diff --git a/src/xslt_ext.c b/src/xslt_ext.c index fec2c21..14e5f23 100644 --- a/src/xslt_ext.c +++ b/src/xslt_ext.c @@ -69,6 +69,6 @@ static void ipcalc_netmask(xmlXPathParserContextPtr ctxt, int nargs) { if (prefix == 0 || prefix > 32) { xsltTransformError(xsltXPathGetTransformContext(ctxt), NULL, NULL, - "ipcalc:netmask: prefix %d not in the range 1 to 32", prefix); + "ipcalc:netmask: prefix %lu not in the range 1 to 32", prefix); goto error; } -- 2.17.1

2 2

[PATCH] dutil_posix: Fix incorrect logic on waitpid
by Eric Blake 06 Sep '18

06 Sep '18

Detected by dcb: dutil_posix.c:225]: (style) Suspicious condition (assignment + comparison); Clarify expression with parentheses. Without the patch, waitret can only ever be 0 or 1, which renders the next line's check for -1 dead code. Signed-off-by: Eric Blake <eblake(a)redhat.com> --- src/dutil_posix.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/dutil_posix.c b/src/dutil_posix.c index 1313514..11ccff6 100644 --- a/src/dutil_posix.c +++ b/src/dutil_posix.c @@ -220,7 +220,7 @@ int run_program(struct netcf *ncf, const char *const *argv, char **output) fclose(outfile); outfile = NULL; - while ((waitret = waitpid(childpid, &exitstatus, 0) == -1) && + while (((waitret = waitpid(childpid, &exitstatus, 0)) == -1) && errno == EINTR) { /* empty loop */ } -- 2.17.1

2 2

[PATCH 0/2] a couple more small changes to the spec file
by Laine Stump 16 Aug '18

16 Aug '18

similar changes were recently made to the libvirt spec file. Laine Stump (2): rpm: simplify applying patches during rpmbuild rpm: remove unused Group: netcf.spec.in | 39 +-------------------------------------- 1 file changed, 1 insertion(+), 38 deletions(-) -- 2.14.4

2 4

[PATCH] remove Requires: bridge-utils, since that package is deprecated
by Laine Stump 16 Aug '18

16 Aug '18

netcf never directly used anything in bridge-utils - this dependency only existed because netcf executes /sbin/ifup (which is in the initscripts package), and in older versions of distros ifup executed brctl, but initscripts lacked a Requires bridge-utils. Signed-off-by: Laine Stump <laine(a)laine.org> --- netcf.spec.in | 4 ---- 1 file changed, 4 deletions(-) diff --git a/netcf.spec.in b/netcf.spec.in index 24a5463..883ed1a 100644 --- a/netcf.spec.in +++ b/netcf.spec.in @@ -92,10 +92,6 @@ developing applications that use %{name}. Summary: Libraries for %{name} Group: System Environment/Libraries -# bridge-utils is needed because /sbin/ifup calls brctl -# if you create a bridge device -Requires: bridge-utils - %description libs The libraries for %{name}. -- 2.14.4

1 0

=?utf-8?q?=5Bnetcf-devel=5D?=(Patch) Fix dead assignment in function modprobed_alias_bond
by Han Han 31 Jul '17

31 Jul '17

commit 41788f6f6a0c3a2c7b30cc2818e118b828ee2a70 (HEAD -> master) Author: Han Han <hhan(a)redhat.com> Date: Mon Jul 31 16:21:27 2017 +0800 Fix dead assignment in function modprobed_alias_bond Drop the line since value stored to 'r' is not used. diff --git a/src/dutil_linux.c b/src/dutil_linux.c index 742153a..ff51cb8 100644 --- a/src/dutil_linux.c +++ b/src/dutil_linux.c @@ -485,7 +485,6 @@ void modprobed_alias_bond(struct netcf *ncf, const char *name) { r = aug_insert(aug, path, "alias", 0); ERR_COND_BAIL(r < 0, ncf, EOTHER); } - r = aug_set(aug, path, name); FREE(path); } -- Han Han Associate Quality Engineer Redhat Email: hhan(a)redhat.com Phone: +861065339333

2 1

[PATCH] Fix a typo in interface.rng
by Yalan Zhang 29 Mar '17

29 Mar '17

There is a typo introduced by netcf-make-start-mode-blah-optional.patch, it is describled in https://bugzilla.redhat.com/show_bug.cgi?id=1436906 --- data/xml/interface.rng | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/data/xml/interface.rng b/data/xml/interface.rng index 46818d1..f9a4693 100644 --- a/data/xml/interface.rng +++ b/data/xml/interface.rng @@ -289,7 +289,7 @@ </choice> </attribute> </element> - </optional + </optional> </define> <!-- -- 1.8.3.1

2 1

[PATCH] make <start mode='blah'/> optional
by Laine Stump 27 Mar '17

27 Mar '17

There's no reason to require it when defining an interface (the config option it corresponds to is optional), and it isn't even output in the status of an interface. --- data/xml/interface.rng | 24 +++++++++++++----------- 1 file changed, 13 insertions(+), 11 deletions(-) diff --git a/data/xml/interface.rng b/data/xml/interface.rng index 9098cc9..46818d1 100644 --- a/data/xml/interface.rng +++ b/data/xml/interface.rng @@ -277,17 +277,19 @@ </define> <define name="startmode"> - <element name="start"> - <attribute name="mode"> - <choice> - <value>onboot</value> - <value>none</value> - <value>hotplug</value> -  - </choice> - </attribute> - </element> + <optional> + <element name="start"> + <attribute name="mode"> + <choice> + <value>onboot</value> + <value>none</value> + <value>hotplug</value> +  + </choice> + </attribute> + </element> + </optional </define> <!-- -- 2.9.3

2 1

[PATCH] dutil_linux: enable netlink peeking to avoid trunc
by Jan Gutter 13 Jan '17

13 Jan '17

When enumerating devices on systems with a large amount of network virtual functions, the netlink receive buffer could be too small and the message gets truncated. This patch enables peeking: libnl will first query the buffer size, expand the receive buffer to the correct size, then receive the full buffer. For a similar issue in libvirt.git, look at commit ID: 8c70d04bab7278c96390a913fa949a17cd3124f9 Reviewed-by: Dinan Gunawardena <dinan.gunawardena(a)netronome.com> Signed-off-by: Jan Gutter <jan.gutter(a)netronome.com> --- src/dutil_linux.c | 1 + 1 file changed, 1 insertion(+) diff --git a/src/dutil_linux.c b/src/dutil_linux.c index f1bf8e0..742153a 100644 --- a/src/dutil_linux.c +++ b/src/dutil_linux.c @@ -687,6 +687,7 @@ int netlink_init(struct netcf *ncf) { goto error; if (nl_connect(ncf->driver->nl_sock, NETLINK_ROUTE) < 0) goto error; + nl_socket_enable_msg_peek(ncf->driver->nl_sock); ncf->driver->link_cache = __rtnl_link_alloc_cache(ncf->driver->nl_sock); if (ncf->driver->link_cache == NULL) -- 2.11.0

2 2

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

netcf-devel