On Mon, 2019-03-18 at 08:08 +0100, Hannes Reinecke wrote:
> On 3/15/19 6:19 PM, Martin Wilck wrote:
> > This can avoid IO while configuring the path prioritizer.
> > The alua prioritizer avoids reading from sysfs for a reason
> > (see commit 7e2f46d3), but this applies only to
On Mon, 2019-03-18 at 08:10 +0100, Hannes Reinecke wrote:
> On 3/15/19 6:19 PM, Martin Wilck wrote:
> > Currently check_rdac() always runs an SG_IO for VPD 0xc9 to check
> > if the storage supports RDAC. This is an extra IO, and may cause
> > annoying error messages on the
On Mon, 2019-03-18 at 10:52 +0100, Martin Wilck wrote:
> On Mon, 2019-03-18 at 08:10 +0100, Hannes Reinecke w
> > *origin? dummy for do_set_from_hwe()?
> > What funky interface is that?
> >
> > Please redesign do_set_from_hwe() so as _not_ to require 'magic
do_set_from_hwe() with statement expression"
(Hannes), renumber subsequent patches accordingly.
- 9/10: cleanup ugly macro usage; do retrieve VPD for "unknown"
SCSI storage devices
Changes v1->v2:
- 6/9: remove one remaining direct access to pp->tpgs
- 9/
Let all alua functions take "const struct path *" as first
argument.
Signed-off-by: Martin Wilck
---
libmultipath/discovery.c | 4 ++--
libmultipath/prioritizers/alua.c | 4 ++--
libmultipath/prioritizers/alua_rtpg.c | 28 +--
li
Signed-off-by: Martin Wilck
---
libmultipath/discovery.c | 2 +-
libmultipath/discovery.h | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/libmultipath/discovery.c b/libmultipath/discovery.c
index 6b4a420b..270dedc9 100644
--- a/libmultipath/discovery.c
+++ b/libmultipath
Hannes Reinecke
Signed-off-by: Martin Wilck
---
libmultipath/propsel.c | 22 +++---
1 file changed, 15 insertions(+), 7 deletions(-)
diff --git a/libmultipath/propsel.c b/libmultipath/propsel.c
index 27474f05..caf55b68 100644
--- a/libmultipath/propsel.c
+++ b/libmultipath/propsel.c
@
We link this already to libmultipath.so. Therefore, no need
to link ti to libprioalua.so, too.
Signed-off-by: Martin Wilck
---
libmultipath/prioritizers/Makefile | 3 ---
1 file changed, 3 deletions(-)
diff --git a/libmultipath/prioritizers/Makefile
b/libmultipath/prioritizers/Makefile
index
This is not the path checker - we don't need to use the
configured checker timeout here. This makes it possible to
call this function without a current (struct config *).
Signed-off-by: Martin Wilck
---
libmultipath/discovery.c | 9 ++---
1 file changed, 6 insertions(+), 3 dele
This can avoid IO while configuring the path prioritizer.
The alua prioritizer avoids reading from sysfs for a reason
(see commit 7e2f46d3), but this applies only to RTPG / STPG,
not to INQUIRY calls.
Cc: steve.schrem...@netapp.com
Cc: NetApp RDAC team
Cc: Hannes Reinecke
Signed-off-by: Martin
ect_prio(), in the case where the alua checker
was statically configured, rather then calling into the alua
code directly, use get_tpgs(), which does all the proper error
checking, and fall back to const prio if it fails.
Signed-off-by: Martin Wilck
---
libmultipath/discovery.c | 10 +++
I got this from Steven.
Cc: steve.schrem...@netapp.com
Cc: NetApp RDAC team
Cc: xose.vazq...@gmail.com
Signed-off-by: Martin Wilck
---
libmultipath/hwtable.c | 20
1 file changed, 20 insertions(+)
diff --git a/libmultipath/hwtable.c b/libmultipath/hwtable.c
index d3a8d9ba
Provide a utility function to retrieve inquiry data from
sysfs, like we do for VPDs.
Signed-off-by: Martin Wilck
---
libmultipath/discovery.c | 23 ++-
libmultipath/discovery.h | 6 --
2 files changed, 22 insertions(+), 7 deletions(-)
diff --git a/libmultipath
"rdac". For "unkown" devices with no checker
configured in either the hwtable or multipath.conf, the VPD call will
be tried.
Cc: steve.schrem...@netapp.com
Cc: NetApp RDAC team
Cc: Hannes Reinecke
Signed-off-by: Martin Wilck
---
libmultipath/propsel.c | 5 +
1 file c
by the "missing udev" logic.
Signed-off-by: Martin Wilck
---
multipathd/main.c | 71 +--
1 file changed, 38 insertions(+), 33 deletions(-)
diff --git a/multipathd/main.c b/multipathd/main.c
index fb520b64..bd26209d 100644
--- a/multipathd/
This is now not needed any more.
Signed-off-by: Martin Wilck
---
libmultipath/structs.h | 1 -
multipathd/main.c | 6 --
2 files changed, 7 deletions(-)
diff --git a/libmultipath/structs.h b/libmultipath/structs.h
index b794b0dc..7879d763 100644
--- a/libmultipath/structs.h
+++ b
nd depend on udev exclusively for WWID generation. I haven't included this
in the current patch set in order not to make it too controversial.
The only purpose for the fallback logic that I could see is to provide
a configuration option to force multipathd to _always_ determine the
WWID by itsel
This option has no effect any more.
Signed-off-by: Martin Wilck
---
libmultipath/config.c | 1 -
libmultipath/config.h | 1 -
libmultipath/dict.c| 18 +++---
multipath/multipath.conf.5 | 8 ++--
4 files changed, 17 insertions(+), 11 deletions(-)
diff --git
On Tue, 2019-03-19 at 12:11 -0500, Benjamin Marzinski wrote:
> On Mon, Mar 18, 2019 at 01:12:32PM +0100, Martin Wilck wrote:
> > Hi Ben, hi Christophe,
> >
> > after reviewing Ben's last patch set, I've been pondering over the
> > handling of changed WWIDs,
On Tue, 2019-03-19 at 12:15 -0500, Benjamin Marzinski wrote:
> On Fri, Mar 15, 2019 at 12:49:41PM +0100, Martin Wilck wrote:
> > On Fri, 2019-03-08 at 17:12 -0600, Benjamin Marzinski wrote:
> > > If multipath can't get the uid for NVMe devices from udev, it can
> >
On Tue, 2019-03-19 at 19:44 +0100, Martin Wilck wrote:
> On Tue, 2019-03-19 at 12:11 -0500, Benjamin Marzinski wrote:
> > On Mon, Mar 18, 2019 at 01:12:32PM +0100, Martin Wilck wrote:
> > >
> > > Note also that if a "reconfigure" was carried out in the prese
ed if we get a
non-zero ID_SERIAL and at the same time none of the required
properties, e.g. ID_WWN. IMO, if this is the case, we can be certain
that scsi_id did *not* fail - after all, it was able to obtain
ID_SERIAL. OTOH, if neither ID_SERIAL nor ID_WWN is set, failure to
access the device is likely. Thus the solution here is simple: We
should apply "blacklisting by missing property" *only* if ID_SERIAL is
set, but ID_WWN is not.
Martin
--
Dr. Martin Wilck , Tel. +49 (0)911 74053 2107
SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel
S device flag if they do
support the DI VPD.
OPEN-V devices are a notorious example. But meanwhile the blist flags
are in place for them.
Regards
Martin
--
Dr. Martin Wilck , Tel. +49 (0)911 74053 2107
SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel
.dev_loss = MAX_DEV_LOSS_TMO,
> },
> {
I'm not quite getting the benefit of adding this comment without
further explanation what Peer Persistence is. Also, IMO, setting
dev_loss_tmo to a higher value than the default (contrary to lowering
it) doesn't r
it's
used much, if at all.
Even if the "least pending" selector was still needed, the patch would
need updating for current upstream, to accomodate the changes from
9659f811446a "dm mpath: push path selector locking down to path
selectors", and possibly other changes in th
On Sat, 2019-03-30 at 15:43 +0100, Xose Vazquez Perez wrote:
> Blacklist "Universal Xport" LUN, it's used for in-band storage array
> management.
> And add it to the rdac dh family.
>
> Cc: Martin Wilck
> Cc: Hannes Reinecke
> Cc: NetApp RDAC team
>
On Sat, 2019-03-30 at 01:05 -0500, Benjamin Marzinski wrote:
> Signed-off-by: Benjamin Marzinski
> ---
Reviewed-by: Martin Wilck
--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel
condlog(1, "%s: failed to get sysfs
> uid: %s",
> - pp->dev, strerror(-len));
> - len = get_vpd_sgio(pp->fd, 0x83, pp-
> >wwid,
> + ssize_t len = -1;
> +
> + if (pp->bus
and
> remembering the "changed WWID" state, this patch simply removes and
> re-adds the path.
>
> This is patch is heavily based on the previous patch of the same name
> by Martin Wilck.
>
> Signed-off-by: Benjamin Marzinski
> ---
Reviewed-by: Martin Wil
+1846,9 @@ get_uid (struct path * pp, int path_state,
> struct udev_device *udev)
> len = get_vpd_uid(pp);
> origin = "sysfs";
On the premise that "udev rules", maybe we should also remove the above
code, which is nothing but yet ano
> the default verbosity.
>
> Signed-off-by: Benjamin Marzinski
> ---
Reviewed-by: Martin Wilck
--
Dr. Martin Wilck , Tel. +49 (0)911 74053 2107
SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
--
dm-devel mailing list
dm-devel@redha
t; return
> value if it doesn't attempt to fallback.
>
> Signed-off-by: Benjamin Marzinski
Reviewed-by: Martin Wilck
--
Dr. Martin Wilck , Tel. +49 (0)911 74053 2107
SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel
Hi Ben,
On Thu, 2019-04-04 at 17:03 -0500, Benjamin Marzinski wrote:
> From: Igor Gnatenko
>
> Reviewed-by: Benjamin Marzinski
> Signed-off-by: Igor Gnatenko
I'd submitted a different patch last month:
"multipath-tools: Build: properly parse systemd's version"
https://www.redhat.com/archives
tipathd: fix daemon not really shutdown
Dominique Leuenberger (1):
multipath-tools: Build: properly parse systemd's version
Martin Wilck (1):
multipathd: protect all access to running_state
Makefile.inc | 9 +--
multipathd/main.c | 60 ++---
ble
are protected by config_lock.
The patch also replaces "if" with "while" in several places where the
code was supposed to wait until a certain state was reached. It's
important that DAEMON_SHUTDOWN terminates all loops of this kind.
Reviewed-by: Benjamin Marzinski
Signed-off-
r "-DUSE_SYSTEMD=241 (241)" results in shell errors.
Try to retrieve the version from pkg-config instead, and if that fails,
discard anything after the first number in "systemctl --version" output.
Acked-by: Benjamin Marzinski
Acked-by: Xose Vazquez Perez
Signed-off-by: Martin W
ove this really happened, so we need add judgement
here too.
(2) [this part removed by mwilck]
Signed-off-by: Chongyun Wu
Signed-off-by: Martin Wilck
Reviewed-by: Benjamin Marzinski
---
multipathd/main.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/multipathd/main.c b/mu
h are not UP or GHOST are ignored anyway, so this
change will not cause changes wrt PG priorities or PG ordering.
Signed-off-by: Martin Wilck
Cc: Hannes Reinecke
---
libmultipath/discovery.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/libmultipath/discovery.c b/libmultipath/discovery.c
o only printed at level 4.
Decrease the verbosity of this message.
Signed-off-by: Martin Wilck
---
libmultipath/prioritizers/ana.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libmultipath/prioritizers/ana.c b/libmultipath/prioritizers/ana.c
index 990d935f..2673d9d9 100644
---
Keep the previous behavior for the case where getting the priority fails
although the path is apparently healthy. This is presumably a very rare
condition, in which it seems actually wrong to preserve the old prio value.
Signed-off-by: Martin Wilck
---
libmultipath/discovery.c | 13 +++--
1 file
pp->priority can be negative, so we we shouldn't compare
it with an unsigned int.
Signed-off-by: Martin Wilck
---
libmultipath/pgpolicies.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libmultipath/pgpolicies.c b/libmultipath/pgpolicies.c
index ac2596ad..660768a
last accessible.
Martin Wilck (4):
libmultipath: group_by_prio: fix signedness bug
Revert "Set priority to '0' for PATH_BLOCKED or PATH_DOWN"
libmultipath: ana prioritizer: decrease log level
libmultipath: get_prio(): don't reset prio for inaccessible path
Hi Christophe,
On Thu, 2019-04-11 at 12:27 +0200, Martin Wilck wrote:
> lest they be forgotten, the patches in this series have been
> previously
> submitted to dm-devel, and got positive review. Please apply at your
> convenience. Ben's latest series "[PATCH v3 00/17] Mi
space" feature in multipath.
One of our customers worked around the problem by adding an
After=systemd-udev-settle.service to multipathd.socket, too. That
way socket activation doesn't happen; it seems to "work" too, but it
wouldn't be my preferred solution.
> Signed-off
Hi Ben and Christophe,
On Thu, 2019-04-18 at 22:30 +0200, Martin Wilck wrote:
> Hi Ben,
>
> > Signed-off-by: Benjamin Marzinski
>
> The patch looks good to me.
>
> Reviewed-by: Martin Wilck
After the holidays, I analyzed this once more, and I've come up with
Since commit d7188fcd "multipathd: start daemon after udev trigger",
multipathd startup is delayed during boot until after "udev settle"
terminates. But "multipath -u" is run by udev workers for storage devices,
and attempts to connect to the multipathd socket. This causes a start job
for multipath
On Wed, 2019-04-24 at 07:48 +0200, Hannes Reinecke wrote:
> On 4/23/19 9:32 PM, Martin Wilck wrote:
> >
> > +/*
> > + * connect to a unix domain socket
> > + */
> > +int mpath_connect(void)
> > +{
> > + int fd = __mpath_connect(0);
> &g
connect() and interpret the
errno appropriately.
This patch reverts most of the changes from commit 8cdf6661 "multipath:
check on multipathd without starting it". Instead, "multipath -u" does
access the socket and start multipath again (which is what we want IMO),
but it is
_info to ->release_clone_rq() for requeue
> path,
> and pass NULL for the others path.
>
> Fixes: 396eaf21ee17 ("blk-mq: improve DM's blk-mq IO merging via
> blk_insert_cloned_request feedback")
> Cc: Ming Lei
> Cc: Martin Wilck
> Cc: Mikulas Patocka
&
On Wed, 2019-04-24 at 11:07 +0200, Martin Wilck wrote:
> Since commit d7188fcd "multipathd: start daemon after udev trigger",
> multipathd startup is delayed during boot until after "udev settle"
> terminates. But "multipath -u" is run by udev workers f
On Sat, 2019-04-27 at 18:54 +0800, Yufen Yu wrote:
> If multipath does not set flag MPATHF_RETAIN_ATTACHED_HW_HANDLER,
> parse_path() would forget to free attached_handler_name when
> successfully exit, resulting in memory leak.
I believe this should rather be done in parse_path(), as
setup_scsi_d
r flag is not set on the map. As
setup_scsi_sh properly "steals" the pointer by nullifying it,
freeing it unconditionally in parse_path() is safe.
Fixes: b592211c33f7 "dm mpath: fix attached_handler_name leak and
dangling hw_handler_name pointer"
Reported-by: Yufen Yu
Signed-
ng, it doesn't get bogged down with passing
> flags
> (and overloading their use with unnatural new meanings). Please let
> me
> know if you see any issues:
>
> https://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm.git/commit/?h=dm-5.2&id=5de719e3d01b4abe0de0d7b857148a880ff2a90b
Nice and simple. Looks good to me.
Reviewed-by: Martin Wilck
--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel
r getting the
expected response for other commands. To avoid this side effect, the client
handling would need to be rewritten such that the uxlsnr thread would have
a means to "know" which commands require the configuration stage to
complete and which do not.
Signed-off-by: Martin Wil
r getting the
expected response for other commands. To avoid this side effect, the client
handling would need to be rewritten such that the uxlsnr thread would have
a means to "know" which commands require the configuration stage to
complete and which do not.
v2: Removed an unrelated,
On Thu, 2019-05-02 at 23:49 +0200, Xose Vazquez Perez wrote:
> Cc: Christophe Varoqui
> Cc: DM-DEVEL ML
> Signed-off-by: Xose Vazquez Perez
> ---
> mpathpersist/main.c | 17 ++---
> 1 file changed, 10 insertions(+), 7 deletions(-)
Reviewed-by: Martin Wilck
-
On Thu, 2019-05-02 at 21:24 +0200, Xose Vazquez Perez wrote:
> Cc: Christophe Varoqui
> Cc: DM-DEVEL ML
> Signed-off-by: Xose Vazquez Perez
> ---
> multipath/main.c | 35 ++-
> 1 file changed, 22 insertions(+), 13 deletions(-)
Reviewed-by: Mar
+++
> 1 file changed, 7 insertions(+)
Reviewed-by: Martin Wilck
--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel
On Mon, 2019-04-29 at 16:38 -0500, Benjamin Marzinski wrote:
> On Thu, Apr 25, 2019 at 09:33:03PM +0200, Martin Wilck wrote:
> >
> > @Ben, I'd be grateful if you could try it (or have the user try it)
> > in your problem case as well.
>
> Unfortunately, I don'
As pointed out by Ben Marzinski, my previous patch ebbb56f2 doesn't do what
it pretends to do.
Fixes: ebbb56f2 "libmultipath: get_prio(): don't reset prio for inaccessible
paths"
---
libmultipath/discovery.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/libmultipath/dis
On Fri, 2019-05-17 at 11:14 -0500, Benjamin Marzinski wrote:
> If clock_gettime() fails, and multipathd can't figure out when it
> should
> time out, it should just default to assuming that it has already
> timed
> out. Found by coverity.
>
> Signed-off-by: Benjamin Marzinski
> ---
> libmultipat
On Fri, 2019-05-17 at 11:14 -0500, Benjamin Marzinski wrote:
> These are fixes to some issues that coverity pointed out.
>
> Benjamin Marzinski (4):
> libmultipath: handle clock_gettime failures in tur checker
> kpartx: fail if dup() of dasd file descriptor fails
> multipathd: fix REALLOC_RE
mpath_presistent_reserve_{in,out} share a lot of common code
for initial data structure initialization (discovery) and teardown.
Factor this code out into mpath_persistent_reserve_init_vecs()
(global data structure initialization),
mpath_persistent_reserve_free_vecs (global teardown) and mpath_get_
We send our PR commands to every active path, regardless of priority.
Thus we can save the effort to obtain priorities.
---
libmpathpersist/mpath_persist.c | 7 ---
1 file changed, 7 deletions(-)
diff --git a/libmpathpersist/mpath_persist.c b/libmpathpersist/mpath_persist.c
index ce72da67..59
Simplify code a bit.
---
mpathpersist/main.c | 11 ++-
1 file changed, 2 insertions(+), 9 deletions(-)
diff --git a/mpathpersist/main.c b/mpathpersist/main.c
index c1a6d3c8..b204647f 100644
--- a/mpathpersist/main.c
+++ b/mpathpersist/main.c
@@ -154,7 +154,7 @@ static int do_batch_file(co
n
- allow multiple PR commands in a single run using "batch files",
without re-discovering everything between PR commands
- avoid looking at priorities
I've done some basic testing and it seems to work. It remains to be
seen what the effect on performance really is.
Regard
Add the option --batch-file (-f) to mpathpersist. The option argument
is a text file that is parsed line-by-line. Every line of the file is
interpreted as an additional input line for mpathpersist. The initial
"mpathpersist" on each line is optional. The '#' character denotes
a comment. '#' is only
We now have the possibility to run several PR commands in a single
mpathpersist invocation. Run initialization / discovery and teardown
only once at program invocation / exit.
---
mpathpersist/main.c | 11 ---
1 file changed, 8 insertions(+), 3 deletions(-)
diff --git a/mpathpersist/main.
We will change the data structure initialization to a lazy
approach, where pp->udev isn't necessarily initialized
when get_mpvec() is called. Deal with it.
---
libmpathpersist/mpath_persist.c | 12
1 file changed, 12 insertions(+)
diff --git a/libmpathpersist/mpath_persist.c b/libmpa
This simplifies further changes.
---
mpathpersist/main.c | 11 ++-
1 file changed, 2 insertions(+), 9 deletions(-)
diff --git a/mpathpersist/main.c b/mpathpersist/main.c
index 10cba452..94e89c13 100644
--- a/mpathpersist/main.c
+++ b/mpathpersist/main.c
@@ -265,7 +265,6 @@ int main (int a
ity.
Not doing this for libdmmp here, as it has it's own error checking
and doesn't use headers from libmultipath.
Signed-off-by: Martin Wilck
---
libmultipath/checkers/tur.c | 6 +++---
libmultipath/time-util.c| 8
libmultipath/time-util.h| 1 +
multipa
tering the current
key, and the CLEAR service action. Fix formatting with longer text in the
description lines (the .TP formatting would produce bad results if the
first line needs to be broken).
Signed-off-by: Martin Wilck
---
mpathpersist/mpathpersist.8 | 43 +--
On Mon, 2019-05-20 at 22:09 +0200, Bart Van Assche wrote:
> On 5/20/19 6:30 PM, Martin Wilck wrote:
> > +void get_monotonic_time(struct timespec *res)
> > +{
> > + struct timespec ts;
> > +
> > + assert(clock_gettime(CLOCK_MONOTONIC, &ts) == 0);
>
ity.
Not doing this for libdmmp here, as it has it's own error checking
and doesn't use headers from libmultipath.
v2: Fix mistake that with -DNDEBUG, clock_gettime wouldn't be called
at all (Bart van Assche).
Signed-off-by: Martin Wilck
---
libmultipath/checkers/tur.c | 6 ++
ntry counting from 1, not 0.
>
> Signed-off-by: Benjamin Marzinski
> ---
> libmultipath/vector.h | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
Reviewed-by: Martin Wilck
Let's add that there was only one caller that actually used
"i"("multipath -W&quo
pathpersist.8: fix examples in man page" which I'd sent
separately before
Martin Wilck (9):
mpathpersist: call usage() just once on return
mpathpersist: add option --batch-file (-f)
mpathpersist: no need to treat error close() as fatal
libmpathpersist: updatepaths: deal with mi
ialization procedure. In get_mpvec(),
we don't delete non-matching maps any more, because we way want to
act on them later on.
Signed-off-by: Martin Wilck
---
libmpathpersist/mpath_persist.c | 231 ++--
libmpathpersist/mpath_persist.h | 40 ++
2 files changed,
We send our PR commands to every active path, regardless of priority.
Thus we can save the effort to obtain priorities.
Signed-off-by: Martin Wilck
---
libmpathpersist/mpath_persist.c | 7 ---
1 file changed, 7 deletions(-)
diff --git a/libmpathpersist/mpath_persist.c b/libmpathpersist
This simplifies further changes.
Signed-off-by: Martin Wilck
---
mpathpersist/main.c | 11 ++-
1 file changed, 2 insertions(+), 9 deletions(-)
diff --git a/mpathpersist/main.c b/mpathpersist/main.c
index 10cba452..94e89c13 100644
--- a/mpathpersist/main.c
+++ b/mpathpersist/main.c
Signed-off-by: Martin Wilck
---
mpathpersist/mpathpersist.8 | 78 -
1 file changed, 77 insertions(+), 1 deletion(-)
diff --git a/mpathpersist/mpathpersist.8 b/mpathpersist/mpathpersist.8
index cd602e40..882043ae 100644
--- a/mpathpersist/mpathpersist.8
+++ b
tering the current
key, and the CLEAR service action. Fix formatting with longer text in the
description lines (the .TP formatting would produce bad results if the
first line needs to be broken).
Signed-off-by: Martin Wilck
---
mpathpersist/mpathpersist.8 | 43 +--
If a command
fails, the batch processing is not aborted. The return status of
mpathpersist is 0 if all commands succeeded, and the status of the
first failed command otherwise.
Signed-off-by: Martin Wilck
---
mpathpersist/main.c | 204
mpathpersist/m
Simplify code a bit.
Signed-off-by: Martin Wilck
---
mpathpersist/main.c | 11 ++-
1 file changed, 2 insertions(+), 9 deletions(-)
diff --git a/mpathpersist/main.c b/mpathpersist/main.c
index 2e19f29f..6df9762d 100644
--- a/mpathpersist/main.c
+++ b/mpathpersist/main.c
@@ -154,7 +154,7
We will change the data structure initialization to a lazy
approach, where pp->udev isn't necessarily initialized
when get_mpvec() is called. Deal with it.
Signed-off-by: Martin Wilck
---
libmpathpersist/mpath_persist.c | 12
1 file changed, 12 insertions(+)
diff
We now have the possibility to run several PR commands in a single
mpathpersist invocation. Run initialization / discovery and teardown
only once at program invocation / exit.
Signed-off-by: Martin Wilck
---
mpathpersist/main.c | 11 ---
1 file changed, 8 insertions(+), 3 deletions
> Signed-off-by: Benjamin Marzinski
> ---
> libmultipath/discovery.c | 3 ++-
> multipathd/cli_handlers.c | 2 +-
> 2 files changed, 3 insertions(+), 2 deletions(-)
Reviewed-by: Martin Wilck
--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel
));
| ^~~~
gcc's stringop checker expects that the size argument of strncat() is derived
from the destination, not source, side.
See
https://developers.redhat.com/blog/2018/05/24/detecting-string-truncation-with-gcc-8/
Fix typo in error message along the way.
Signed-off-by: M
merge_words() is always called with space = 1. Remove the argument.
Signed-off-by: Martin Wilck
---
libmultipath/dmparser.c | 18 +++---
1 file changed, 7 insertions(+), 11 deletions(-)
diff --git a/libmultipath/dmparser.c b/libmultipath/dmparser.c
index ac13ec06..04f675c1 100644
patches 14, 18, 21, 22, 24).
Reviews and comments welcome.
Regards,
Martin
Martin Wilck (30):
kpartx: dasd: fix -Waddress-of-packed-member warning from gcc9
libmultipath: fix gcc -Wstringop-truncation warning in set_value()
libmultipath: remove 'space' argument to merge_words()
This is nothing but a reimplementation of strdup(), and it causes gcc 9
warnings. Remove it.
Signed-off-by: Martin Wilck
---
libmultipath/debug.c| 1 +
libmultipath/defaults.c | 17 -
libmultipath/defaults.h | 9 -
3 files changed, 9 insertions(+), 18 deletions
’:
dmparser.c:41:19: note: length computed here
41 | strncpy(p, word, strlen(word) + 1);
| ^~~~
Signed-off-by: Martin Wilck
---
libmultipath/dmparser.c | 15 ++-
1 file changed, 6 insertions(+), 9 deletions(-)
diff --git a/libmultipath/dmparser.c b
More often than not, this means replacing strncpy() by strlcpy().
Signed-off-by: Martin Wilck
---
libmpathpersist/mpath_persist.c | 8
libmultipath/configure.c| 7 ---
libmultipath/discovery.c| 12 ++--
libmultipath/dmparser.c | 20
Signed-off-by: Martin Wilck
---
libmultipath/discovery.c | 49 ++--
1 file changed, 32 insertions(+), 17 deletions(-)
diff --git a/libmultipath/discovery.c b/libmultipath/discovery.c
index 45c2931e..e484ec8d 100644
--- a/libmultipath/discovery.c
+++ b
The length of the uuid field in libdm is DM_UUID_LEN, which happens
to be one byte more than our WWID_SIZE. Handle this cleanly.
Signed-off-by: Martin Wilck
---
libmultipath/devmapper.c | 28 +---
libmultipath/devmapper.h | 2 +-
libmultipath/wwids.c | 3
Signed-off-by: Martin Wilck
---
libmpathcmd/mpath_cmd.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libmpathcmd/mpath_cmd.c b/libmpathcmd/mpath_cmd.c
index b681311b..05b69316 100644
--- a/libmpathcmd/mpath_cmd.c
+++ b/libmpathcmd/mpath_cmd.c
@@ -104,7 +104,7 @@ int
We set out[len] = '\0' later, thus we should set len to no more then
out_len - 1.
Fixes: 756ef73b7197 "Separate out vpd parsing functions"
Signed-off-by: Martin Wilck
---
libmultipath/discovery.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libmul
uot;Read wwid from sysfs vpg_pg83 attribute"
Signed-off-by: Martin Wilck
---
libmultipath/discovery.c | 11 +--
1 file changed, 9 insertions(+), 2 deletions(-)
diff --git a/libmultipath/discovery.c b/libmultipath/discovery.c
index 407e64a0..f360e306 100644
--- a/libmultipath/disc
As we're using strlcpy quite a bit, make sure it works as designed.
Signed-off-by: Martin Wilck
---
tests/util.c | 142 +++
1 file changed, 142 insertions(+)
diff --git a/tests/util.c b/tests/util.c
index e6d4b9ab..4e04a480 100644
--- a/
Signed-off-by: Martin Wilck
---
tests/hwtable.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/hwtable.c b/tests/hwtable.c
index 13da1d99..f436f52d 100644
--- a/tests/hwtable.c
+++ b/tests/hwtable.c
@@ -457,7 +457,7 @@ static void replicate_config(const struct
Signed-off-by: Martin Wilck
---
tests/uevent.c | 18 ++
1 file changed, 18 insertions(+)
diff --git a/tests/uevent.c b/tests/uevent.c
index 215d97ad..f4afd9b4 100644
--- a/tests/uevent.c
+++ b/tests/uevent.c
@@ -95,6 +95,23 @@ static void test_ro_good(void **state
301 - 400 of 2237 matches
Mail list logo
