| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
| |
Using the log_analyzer script analyze the bricks and client logs, which
tells what functions are called how much times. Thus excessive logging can
be stopped.
Change-Id: I74ee71b29865612332603a2ffc7fb5c7055c921d
Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com>
|
|
|
|
|
| |
Change-Id: Id67a4096d03394499241bcc01154b60bcc0bc5a9
Signed-off-by: Anush Shetty <ashetty@redhat.com>
|
|
|
|
|
|
|
|
|
|
| |
default file
Fetch the configuration file for dbench from the known location instead of
depending upon the default path.
Change-Id: Id7fd39a4cf7765c22500e2771c2cd072e8f04d9c
Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com>
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
*Adding check so that program does not crash if
getcwd fails.
*Defining macro so as to remove code duplication.
Change-Id: Ia884b3fa018d4dfeec56c6f20de413703c74fad7
Signed-off-by: Rahul C S <rahulcs@redhat.com>
|
| |
| |
| |
| |
| |
| |
| | |
kernel_compile
Change-Id: I1f84a1fed05c15c5258e66d39d87fb59e952a111
Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com>
|
|/
|
|
|
| |
Change-Id: Idd837357f61aba66491ef4f323fdc3a73d8842b4
Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com>
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
sparse_file_write.c: This program takes the filename and the size as
the argument from the user (default size 5GB) and
creates the sparse file with the name and the size
provided.
sparse_file_alt.c: This program will create a sparse file with alternate
empty blocks and data blocks. Takes the path as the
argument
Change-Id: I8dfc95c933167fefc53fb0b665f234997aa0c68f
Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com>
|
| |
| |
| |
| |
| |
| |
| | |
*Also confessing blunders on fs_mark sanity. :(
Signed-off-by: Rahul C S <rahulcs@redhat.com>
Change-Id: I7027f90879d3e9ba9f95cc79aca726b7ab9a6d8f
|
|\ \ |
|
| | |
| | |
| | |
| | |
| | | |
Change-Id: I192d2b8eaf476d1bb7d0a1c0365a1a5799ac577f
Signed-off-by: Rahul C S <rahulcs@redhat.com>
|
|\ \ \ |
|
| | |/
| |/|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Thanks to Avati <aavati@redhat.com> for the patch
Helps creating .glusterfs directory hierarchy
structure in the backend. Could be useful for the
client as well.
Signed-off-by: Rahul C S <rahulcs@redhat.com>
Change-Id: Ia540c4b6a3e4ea3b05aa7de216ff6c3734eb5792
|
| | |
| | |
| | |
| | |
| | | |
Change-Id: I77f5e82a6a04166dd200173cffa097d7b74cff9c
Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Also make changes to take mount type as argument
Change-Id: Ie13d014dbacf5063f0c505d733c30a999c56deb7
Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com>
|
|/ /
| |
| |
| |
| |
| |
| |
| |
| | |
*Changes to exit program if peer probe is not successful.
*Flush firewall rules before doing peer probe.
*White-space cleanup in create_gluster_vol script.
Change-Id: Ia44fd76fa1e72160daa80ef7206cffc2bd267887
Signed-off-by: Rahul C S <rahulcs@redhat.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
With changes such as gfid based backend in glusterfs, we have to ignore some
hidden directories from the backend (which creates the hidden directories for
better management of the volume). So this patch takes care of taking the list
of ignored directories from user as an input argument. Those directories which
are present in the path given by the user (of which arequal is calculating the
checksum), are not considered while crawling the path.
Change-Id: If6b2209088300c5c59a8b0fb74e854f3ce030571
Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com>
|
| |
| |
| |
| |
| |
| |
| |
| | |
Adding arequal to the git, which can be used to calculate the checksum
of path given to it as an argument
Change-Id: Iefd283511e6ff2f3bc1359e718d1922506d7a24a
Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com>
|
| |
| |
| |
| |
| | |
Change-Id: I111eb271116ce36dafd121fa7cee176539623eea
Signed-off-by: Rahul C S <rahulcs@redhat.com>
|
|\ \ |
|
| |/
| |
| |
| |
| |
| |
| |
| |
| | |
Till now this script allowed running all the tools (or exit for the
first failure). Running individual tools was not possible. Now
individual tools can be given as an argument to run.
Change-Id: Ia133943667ad4dc30956f4a021f2edfc30458cee
Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com>
|
|\ \ |
|
| |/
| |
| |
| |
| | |
Change-Id: Ia7dbd82e9bb2e5e65e9345234ce34f8518a091ad
Signed-off-by: Rahul C S <rahulcs@redhat.com>
|
|/
|
|
|
|
|
|
|
|
| |
Suppose due to some error (or due to the change in the filename
pattern of the statedump files), if statedump files were not copied,
then do not error out. Instead log a message saying statedump was
not taken, and continue with the tests.
Change-Id: I6d7679f78461be7eac2c2ee8c7939c08c734ff1c
Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com>
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
geo-replication and quota were not enabled during the tests, and if
they were needed they had to be hardcoded in the script. Instead let the
user who is running the script decide whether to enable any of them (or
both of them), by giving arguments at the time of executing the script.
Change-Id: I228eab1b9df4a4c261fb48c33618b9b56801a3d9
Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com>
|
|/
|
|
|
|
|
|
|
|
|
| |
This patch provides an option for the user to give the number of threds
as an argument. Before, ping_pong functionlity (i.e. locking the byte range)
was done in a separate single thread. With this patch depending upon the
input argument, multiple threads will be spawned all of which will try
to do the locking using the same fd.
Change-Id: Ib68bef407d75351664a76e81474814bd703d285c
Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com>
|
|
|
|
|
|
|
|
| |
Changes to take the working directory and the time duration for
which the test has to run from the user.
Change-Id: I2d56b0b0b640675dd52f4f3a6f17229e3cdf5d1a
Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com>
|
|
|
|
|
|
|
|
|
| |
Since iozone and bonnie were part of the performance script before
they were removed from the nightly sanity scripts. Now they are
again added to the sanity script.
Change-Id: Iefb502dda905b93987d5180af9ac474925c9407b
Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com>
|
|\
| |
| |
| | |
address"
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Currently the stack address of the structure which holds information about
the number of locks, fd etc is passed to the thread doing the locking.
Instead of sending the stack address, allocate memory from the heap and
use that address to send to the other thread.
Change-Id: I417e22aee0eed1fa921982e78239147553886786
Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com>
|
|\ \ |
|
| |/
| |
| |
| |
| | |
Change-Id: Ief7b6f884cf08321343e03758eeedb312b1fe799
Signed-off-by: Rahul C S <rahulcs@redhat.com>
|
|\ \ |
|
| |/
| |
| |
| |
| |
| |
| |
| | |
Compile glusterfs with gcov compatible flags and generate the report
of how much code has been covered by the tests.
Change-Id: I1d3c733cc93625c4f14acc0dea382a20a3b35672
Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com>
|
|/
|
|
|
| |
Change-Id: I6004e4066fab30e14716ef6475c1281270974d2b
Signed-off-by: M S Vishwanath Bhat <vishwanath@gluster.com>
|
|
|
|
|
| |
Change-Id: Ia15b0a164a75385527103c7723ef12cf2fff4f55
Signed-off-by: Vijay Bellur <vijay@gluster.com>
|
|
|
|
|
|
|
| |
Some benchmarking related changes in threaded-io to print total
attempted and successful syscalls executed.
Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com>
|
|
|
|
|
|
|
| |
Some minor changes in the probe script (change in the function names)
and detach script is added to detach the peers in the cluster.
Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com>
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
parallely on multiple machines
installs the glusterfs rpm of specified version
(can upgrade too depending upon the arguments), on
multiple machines parallely. And also some changes to the rpm installing
script helper_scripts/rpm_download_install.
|
| |
| |
| |
| |
| |
| |
| | |
glusterfs rpms
This script not only installs the qa releases but also installs the main releases also.
Can be even upgraded by giving option "yes".
|
| |
| |
| |
| |
| | |
scripts which uninstalls glusterfs on the local machine and on a set
of machines.
|
| |
| |
| |
| |
| |
| |
| |
| | |
some offset
This program creats a file if it does not exist (filename given as an argument), opens
it truncates it to the size mentioned in the argument (default 5GB), then writes some
data to some offset. Thus in general creates a sparse file and write into it.
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
new statedump files
Previously statedump files were located in /tmp/with the name glusterfs.<pid>
Now (glusterfs-3.3.0qa12 onwards) its not like that. Brick statedumps are saved
in the pattern "brick-path.<pid>.dump" and client statedumps are saved in the
pattern glusterdump.<pid>.dump. Changes are done with this commit to make nightly
sanity script aware of that.
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
simultaneously on multiple machines
This scripts looks into a file to know the IP address of the machines
where glusterfs has to be installed. Then executes the download_and_install.sh
script on all the machines. It is assumed that download_and_install.sh
script is available on all the machines. glusterfs version is given
as the argument.
|
|/ |
|
|
|
|
|
|
| |
Do not print the git diff of the applied patches in the message.
Instead copy the patches diretory to /tmp and save that directory
in the results directory everyday.
|
|
|
|
| |
build with DEBUG
|