summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Land part 2 of clang-format changesGluster Ant2018-09-12490-439790/+426792
| | | | | Change-Id: Ia84cc24c8924e6d22d02ac15f611c10e26db99b4 Signed-off-by: Nigel Babu <nigelb@redhat.com>
* Land clang-format changesGluster Ant2018-09-12364-31211/+29701
| | | | Change-Id: I6f5d8140a06f3c1b2d196849299f8d483028d33b
* clang-format: add the config fileAmar Tumballi2018-09-125-1/+151
| | | | | | | | Also update the required documents and scripts to enable clang-format Change-Id: I73aae6db06c2f732a1779d59a73bc05e28beafba updates: bz#1564149 Signed-off-by: Amar Tumballi <amarts@redhat.com>
* doc: make developer-index.md as READMEAmar Tumballi2018-09-121-0/+0
| | | | | | | | | | This helps to traverse the docs well as README.md is rendered by default in github pages, and would help to provide better indexed document for new users. updates: bz#1193929 Change-Id: I4130dfd16aed924ecab17f96326cedb86892d6a6 Signed-off-by: Amar Tumballi <amarts@redhat.com>
* dht: Use snprintf instead of strncpyN Balachandran2018-09-121-9/+20
| | | | | | | | | | | The recent changes to use malloc instead of calloc left the new_name and new_path non-null terminated. We now use snprintf instead of strncpy to fix this. Change-Id: I1a31701ca9447efde38921be0ba2c73cde2e7976 fixes: bz#1626346 Signed-off-by: N Balachandran <nbalacha@redhat.com>
* doc: add coding-standard and commit-msg link in READMEAmar Tumballi2018-09-111-0/+8
| | | | | | | | | Having the links in README makes people follow it if they browse through github. Updates: bz#1193929 Change-Id: I580332a41d9b52858aa9796a74a563ff57a60b4b Signed-off-by: Amar Tumballi <amarts@redhat.com>
* mount/fuse: convert ENOENT to ESTALE in open(dir)_resumeRaghavendra G2018-09-111-0/+9
| | | | | | | | | | | | | | | | | | | | | | This patch is continuation of commit fb4b914ce84bc83a5f418719c5ba7c25689a9251. <snip> mount/fuse: never fail open(dir) with ENOENT open(dir) being an operation on inode should never fail with ENOENT. If gfid is not present, the appropriate error is ESTALE. This will enable kernel to retry open after a revalidate lookup. </snip> Earlier commit failed to fix codepath where error response is sent back on gfid resolution failures in fuse_open(dir)_resume. Current patch completes that work Change-Id: Ia07e3cece404811703c8cfbac9b402ca5fe98c1e Signed-off-by: Raghavendra G <rgowdapp@redhat.com> updates: bz#1627620
* mgmt xlators: store boolean fields using integerYaniv Kaul2018-09-112-8/+11
| | | | | | | | | | | | | | | | | | | | Surprisingly, there is not set_boolean() as there is a get_boolean() In fact, it is stored as an INT dictionary type. In some occasions it was stored using a string, and this caused errors such as: key gfproxy-server, integer type asked, has string type [Invalid argument] I've fixed what I saw in some logs, I'm sure there are more. The CORRECT fix is to create a boolean set and use it, but this requires a bit more work. I'll see if I can do it later on. Only compile-tested! updates: bz#1193929 Signed-off-by: Yaniv Kaul <ykaul@redhat.com> Change-Id: I45fd0c7a0824b2f42b8ce510296c9dfa4f32ad66
* glusterd: NULL pointer dereferencing clang fixHarpreet Lalwani2018-09-111-2/+1
| | | | | | | | | | | problem: NULL point dereferencing solution: Adding a conditional statement before and then dereferencing it. Updates: bz#1622665 Change-Id: I562ca90aebf2a4882cfea10114a90364d9ef1996 Signed-off-by: Harpreet Lalwani <hlalwani@redhat.com>
* misc: fix misc. shebangsKaleb S. KEITHLEY2018-09-118-8/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * One #!/usr/bin/env python and three #!/usr/bin/python were overlooked in all the other python fixups. Ugh. * Two new python files missed the memo about #!/usr/bin/python3. * One #!/usr/bin/env bash. Various distribution packaging policies have strong wording about the use of #!/usr/bin/env ... Note: this patch does not change the use of #!/usr/bin/env bash in the two files extras/{clang-checker.sh,check_goto.pl} as these are not included in any packages. (Although I'm not actually sure why anyone would ever use '/usr/bin/env {sh,bash}' as I'm not aware of any version-specific differences like there are with, e.g., python.) * One #!/usr/bin/bash. On Fedora and CentOS > 6, /bin is a symlink to /usr/bin, so it makes little difference. But Debian & Ubuntu still have separate /bin and /usr/bin; and sh and bash are in /bin, not /usr/bin. (Historically, in BSD and SYSV Unix it was /bin/sh.) Note: Fedora and CentOS package build runs a script that converts all /bin/sh and /bin/bash to /usr/bin/sh and /usr/bin/bash. Change-Id: I9171265829af78dd0cd7622c22b56d22179ff8a3 updates: bz#1193929 Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com>
* rpm: add "--with asan" build optionNiels de Vos2018-09-112-1/+17
| | | | | | | | | | | | RHEL-6 does not have libasan, enabling the "--with asan" option is a no-op there. RHEL-7 has an earlier version of libasan, and that does not have the __asan_init symbol. Test for __asan_report_error in confiure.as instead. Change-Id: I6322e832c5cfbd7d750f5c32c84c28771674ced6 Updates: #492 Signed-off-by: Niels de Vos <ndevos@redhat.com>
* storage/posix: Fix coverity issue - Unchecked return valueAshish Pandey2018-09-111-1/+8
| | | | | | | | | | Fixes CID: 1388886 https://scan6.coverity.com/reports.htm#v42607/p10714/fileInstanceId=85287446&defectInstanceId=25997291&mergedDefectId=1388886 Change-Id: Ic4e558bba7e15d213c07bc31affb2e175ace5502 updates: bz#789278 Signed-off-by: Ashish Pandey <aspandey@redhat.com>
* performance/write-behind: remove the request from wip queue in ↵Raghavendra G2018-09-111-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | wb_fulfill_request The bug is very similar to bz 1379655 and the fix too very similar to commit a8b2a981881221925bb5edfe7bb65b25ad855c04. Before this patch, a request is removed from wip queue only when ref count of request hits 0. Though, wb_fulfill_request does an unref, it need not be the last unref and hence the request may survive in wip queue till the last unref. Let, T1: the time at which wb_fulfill_request is invoked T2: the time at which last unref is done on request Let's consider a case of T2 > T1. In the time window between T1 and T2, any other request (waiter) conflicting with request in liability queue (blocker - basically a write which has been lied) is blocked from winding. If T2 happens to be when wb_do_unwinds is invoked, no further processing of request list happens and "waiter" would get blocked forever. An example imaginary sequence of events is given below: 1. A write request w1 is picked up for winding in __wb_pick_winds and w1 is moved to wip queue. Let's call this invocation of wb_process_queue by wb_writev as PQ1. Note w1 is not unwound. 2. A dependent write (w2) hits write-behind and is unwound followed by a flush (f1) request. Since the liability queue of inode is not empty, w2 and f1 are not picked for unwinding. Let's call the invocation of wb_process_queue by wb_flush as PQ2. Note that invocation of wb_process_queue by w2 doesn't wind w2 instead unwinds it after which we hit PQ2 3. PQ2 continues and picks w1 for fulfilling and invokes wb_fulfill. As part of successful wb_fulfill_cbk, wb_fulfill_request (w1) is invoked. But, w1 is not freed (and hence not removed from wip queue) as w1 is not unwound _yet_ and a ref remains (PQ1 has not invoked wb_do_unwinds _yet_). 4. wb_fulfill_cbk (triggered by PQ2) invokes a wb_process_queue (let's say PQ3). w2 is not picked up for winding in PQ3 as w1 is still in wip queue. At this time, PQ2 and PQ3 are complete. 5. PQ1 continues, unwinds w1 and does last unref on w1 and w1 is freed (and removed from wip queue). Since PQ1 didn't invoke wb_fulfill on any other write requests, there won't be any future codepaths that would invoke wb_process_queue and w2 is stuck forever. This will prevent f2 too and hence close syscall is hung With this fix, w1 is removed from liability queue in step 3 above and PQ3 winds w2 in step 4 (as there are no requests conflicting with w2 in liability queue during execution of PQ3). Once w2 is complete, f1 is resumed. Change-Id: Ia972fad0858dc4abccdc1227cb4d880f85b3b89b Signed-off-by: Raghavendra G <rgowdapp@redhat.com> Fixes: bz#1626787
* cli: Add warning message while converting to replica 2 configurationkarthik-us2018-09-113-15/+59
| | | | | | | | | | | | | | Currently while creating replica 2 volume we display a warning message of ending up in split-brain. But while converting an existing volume from other configuration to replica 2 by add-brick or remove-brick operations we do not show any such messages. With this fix in add-brick and remove-brick cases also we will display the same warning message and prompt for confirmation if the configuration changes to replica 2. Change-Id: Ifc4ed6994a087d2403894f4e743c4eb41633276b fixes: bz#1627044 Signed-off-by: karthik-us <ksubrahm@redhat.com>
* mgmt/glusterd : Fix coverity issueAshish Pandey2018-09-112-6/+13
| | | | | | | | | | | | CID: 727146, 727066 https://scan6.coverity.com/reports.htm#v42607/p10714/fileInstanceId=85393035&defectInstanceId=26034751&mergedDefectId=727146 https://scan6.coverity.com/reports.htm#v42607/p10714/fileInstanceId=85392913&defectInstanceId=26034571&mergedDefectId=727066 updates: bz#789278 Change-Id: Ieaef33829ec88e68690dabce4ea21d2e61dad9f6 Signed-off-by: Ashish Pandey <aspandey@redhat.com>
* glusterd: NULL pointer dereferencing clang fixSheetal Pamecha2018-09-111-1/+2
| | | | | | | | | Added ternary operator to avoid NULL pointer dereferencing Updates: bz#1622665 Change-Id: I4b970176b6b555c2eda1da2848c493e45b1e4217 Signed-off-by: Sheetal Pamecha <sheetal.pamecha08@gmail.com>
* features/snapview-server: remove the ret variableRaghavendra Bhat2018-09-111-4/+2
| | | | | | | | Used only once to store the return value of a function. Change-Id: Ib2b9db847b5f652ce46f220297d9c7c3503eaea2 fixes: #230 Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
* doc: add details on how the commit should be structuredAmar Tumballi2018-09-101-0/+136
| | | | | | | | | | | | | | Detailing how the commit message should be is very important for imposing some structure, and good practises. Also it helps new developers to do the right things. This is mostly copied from the OpenStack commit message reference, as it was overlapping with most of the things we follow in GlusterFS. Made minor changes to make it suitable for Gluster project. Updates: bz#1193929 Change-Id: I4dcd7ff366edfc6c1727a287f32bcf7270c86b11 Signed-off-by: Amar Tumballi <amarts@redhat.com>
* rfc.sh: use 'git log -n1' instead of 'git show'Amar Tumballi2018-09-101-4/+4
| | | | | | | | | | | | | | This is required mainly because if we use any format in a documentation, and it has strings which are looking for commit msg format (like fixes: bz...) in document, that too will be considered in checks, even though it is not part of commit msg. By using git log -n1, we would only check the content of commit msg. Updates: bz#1193929 Change-Id: I498590de6d3daa6be7bcbee1f083ef3fa6ecc96f Signed-off-by: Amar Tumballi <amarts@redhat.com>
* cluster/dht: Create a linkto file if requiredN Balachandran2018-09-101-0/+30
| | | | | | | | | | | Using the dht_filter_loc_subvol_key to create files on specific subvols did not create a linkto file. This can make the file inaccessible as lookup-optimize is now enabled by default. Change-Id: I78add5a31887378a479cb9c746b91678876b0dbe fixes: bz#1626394 Signed-off-by: N Balachandran <nbalacha@redhat.com>
* build: Make building of glupy more obviousAnoop C S2018-09-101-12/+8
| | | | | | | | | | | | | | | | | | | | | | Yet again glupy..!! Core changes: * Provide --disable-glupy(instead of --enable-glupy) since we try to build it by default. * AC_SUBST macro is considered/executed even if it is inside an if condition. As a result GLUPY_SUBDIR was always substituted which in turn would make compilation enter glupy/ translator sources in absence of python{2/3}-devel and fails. Miscellaneous changes: * Remove an explicit echo for printing PYTHON version. * Replace hard coded python version in warning message displayed in absence of python{2/3}-devel. * Redirect pushd and popd output to /dev/null Change-Id: If1ba30a12a8bda5763ef528787fccb2f7946b136 Updates: bz#1193929 Signed-off-by: Anoop C S <anoopcs@redhat.com>
* posix: fix the duplicate macro definitionsAmar Tumballi2018-09-102-8/+0
| | | | | | | | | | the macro 'PATH_SET_TIMESPEC_OR_TIMEVAL' is defined in posix.h, which seems to be good place for it updates: bz#1193929 Change-Id: I521a2a6efeb26891c24637a5f06b1d90c00b1135 Signed-off-by: Amar Tumballi <amarts@redhat.com>
* cluster/dht: optimize readdir for 1xn volsN Balachandran2018-09-101-12/+42
| | | | | | | | | Skip the hashed subvol check for volumes with distribute count of 1. Change-Id: I5703508b54a17c49a217c8a8e09884980705953a fixes: bz#1608175 Signed-off-by: N Balachandran <nbalacha@redhat.com>
* ec-heal: remove a duplicate definition of alloca0Amar Tumballi2018-09-101-1/+0
| | | | | | | | | | the same macro is defined in common-utils.h, which seems to be much better place for the same. Updates: bz#1193929 Change-Id: I409b719c291102136500b955e5827a550142ed96 Signed-off-by: Amar Tumballi <amarts@redhat.com>
* xlators/cluster/dht/src/tier-common.c:move to GF_MALLOC() instead of ↵Yaniv Kaul2018-09-101-1/+1
| | | | | | | | | | | | | | | | | | GF_CALLOC() when It doesn't make sense to calloc (allocate and clear) memory when the code right away fills that memory with data. It may be optimized by the compiler, or have a microscopic performance improvement. Please review carefully, especially for string allocation, with the terminating NULL string. Only compile-tested! Change-Id: I7d38a7d576f6777976fe86e5351a8d95caddbb9c updates: bz#1193929 Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
* Modify log message 'DH ciphers are disabled' from ERROR to INFOOmar Kohl2018-09-101-1/+1
| | | | | | | | | | Per the latest comment in bz#1398237 this message is confusing for users because it suggests an error where none exists. Fixes: bz#1626319 Change-Id: I2f05999da157b11e225bf3d95edb597e964f9923 Signed-off-by: Omar Kohl <omarkohl@gmail.com>
* glusterd: Fix unused value coverity fixSanju Rakonde2018-09-091-1/+1
| | | | | | | | | | Commit 09198e203e has introduced a new coverity with ID 1395635. keylen variable is assigned to some value but stored value is overwritten before it is used. This patch addresses the issue. updates: bz#789278 Change-Id: Ice290dcb9d703cd2131b0f0803436660e670e10a Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
* Some (mgmt) xlators: use dict_{setn|getn|deln|get_int32n|set_int32n|set_strn}Yaniv Kaul2018-09-0925-1718/+2425
| | | | | | | | | | | | | | | | | | | | | In a previous patch (https://review.gluster.org/20769) we've added the key length to be passed to dict_* funcs, to remove the need to strlen() it. This patch moves some xlators to use it. - It also adds dict_get_int32n which was missing. - It also reduces the size of some key variables. They were set to 1024b or PATH_MAX, where sometimes 64 bytes were really enough. Please review carefully: 1. That I did not reduce some the size of the key variables too much. 2. That I did not mix up some keys. Compile-tested only! Change-Id: Ic729baf179f40e8d02bc2350491d4bb9b6934266 updates: bz#1193929 Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
* glusterd: avoid using glusterd's working directory as a brickSanju Rakonde2018-09-082-0/+18
| | | | | | | | | Adding checks for avoiding glusterd's working directory used as a brick for volume creation. fixes: bz#853601 Change-Id: I4b16a05f752e92216aa628f542a4fdbf59b3c669 Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
* multiple xlators (mgmt): strncpy()->sprintf(), reduce strlen()'sYaniv Kaul2018-09-078-268/+482
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | xlators/mgmt/glusterd/src/glusterd-geo-rep.c xlators/mgmt/glusterd/src/glusterd-handshake.c xlators/mgmt/glusterd/src/glusterd-sm.c xlators/mgmt/glusterd/src/glusterd-store.c xlators/mgmt/glusterd/src/glusterd-utils.c xlators/mgmt/glusterd/src/glusterd-volgen.c xlators/mgmt/glusterd/src/glusterd-volume-ops.c xlators/mgmt/glusterd/src/glusterd.c strncpy may not be very efficient for short strings copied into a large buffer: If the length of src is less than n, strncpy() writes additional null bytes to dest to ensure that a total of n bytes are written. Instead, use snprintf(). Try to ensure output is not truncated. Also: - save the result of strlen() and re-use it when possible. - move from strlen to SLEN (sizeof() ) for const strings. Compile-tested only! Change-Id: Ib5d001857236f43e41c4a51b5f48e1a33110aaeb updates: bz#1193929 Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
* glusterd-sm: NULL pointer dereferencing clang fixShwetha K Acharya2018-09-071-0/+1
| | | | | | | | | | Problem: new_event could be NULL. Solution: Added a goto statement to address this issue. Updates: bz#1193929 Change-Id: Id3ce28fc53ad2cc8b9fcb63f7774568d31073b9e Signed-off-by: Shwetha K Acharya <sacharya@redhat.com>
* cluster/dht: Rework the debug xattr to get hashed subvolN Balachandran2018-09-073-28/+87
| | | | | | | | | | | | | | | | | | | The earlier implementation required the file to already exist when trying to get the hashed subvol. The reworked implementation allows a user to get the hashed subvol for any filename, whether it exists or not. Usage: getfattr -n "dht.file.hashed-subvol.<filename>" <parent dir> Eg:To get the hashed subvol for file-1 inside dir-1 getfattr -n "dht.file.hashed-subvol.file-1" /mnt/gluster/dir1 credit: rgowdapp@redhat.com Change-Id: Iae20bd5f56d387ef48c1c0a4ffa9f692866bf739 fixes: bz#1624244 Signed-off-by: N Balachandran <nbalacha@redhat.com>
* cluster/ec: Don't update trusted.ec.version if fop succeedsAshish Pandey2018-09-071-0/+9
| | | | | | | | | | | | | | | | | | | | If a fop has succeeded on all the bricks and trying to release the lock, there is no need to update the version for the file/entry. All it will do is to increase the version from x to x+1 on all the bricks. If this update (x to x+1) fails on some brick, this will indicate that the entry is unhealthy while in realty everything is fine with the entry. Avoiding this update will help to not to send one xattrop at the end of the fops. Which will decrease the chances of entries being in unhealthy state and also improve the performance. Change-Id: Id9fca6bd2991425db6ed7d1f36af27027accb636 fixes: bz#1623759 Signed-off-by: Ashish Pandey <aspandey@redhat.com>
* cluster/ec: Improve logging for some critical error messagesAshish Pandey2018-09-073-14/+55
| | | | | | Change-Id: I037e52a3467467b81a1ba5416317870864060d4d updates: bz#1615703 Signed-off-by: Ashish Pandey <aspandey@redhat.com>
* Quota: Fix for clang issuessanoj-unnikrishnan2018-09-073-5/+6
| | | | | | Updates: bz#1193929 Change-Id: I549f741fbf97ee3fa2d63edacee5b705ef37c49a Signed-off-by: sanoj-unnikrishnan <sunnikri@redhat.com>
* xlators/cluster/dht/src/dht-common.c :move to GF_MALLOC() instead of ↵Yaniv Kaul2018-09-071-2/+2
| | | | | | | | | | | | | | | | | | GF_CALLOC() when It doesn't make sense to calloc (allocate and clear) memory when the code right away fills that memory with data. It may be optimized by the compiler, or have a microscopic performance improvement. Please review carefully, especially for string allocation, with the terminating NULL string (added another byte to ensure it's there). Only compile-tested! Change-Id: Ia5e4f50dfb0c29809c2019fcfd8079507813249e updates: bz#1193929 Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
* glusterfsd/src/glusterfsd.c: Move to GF_MALLOC() instead of GF_CALLOC() when ↵Yaniv Kaul2018-09-071-5/+5
| | | | | | | | | | | | | | | | | | | | | | | | | possible It doesn't make sense to calloc (allocate and clear) memory when the code right away fills that memory with data. It may be optimized by the compiler, or have a microscopic performance improvement. In some cases, also changed allocation size to be sizeof some struct or type instead of a pointer - easier to read. In some cases, removed redundant strlen() calls by saving the result into a variable. 1. Only done for the straightforward cases. There's room for improvement. 2. Please review carefully, especially for string allocation, with the terminating NULL string. Only compile-tested! updates: bz#1193929 Signed-off-by: Yaniv Kaul <ykaul@redhat.com> Change-Id: Iaed86fcc909022c5158c3e08a9106b1110b9df0a
* multiple xlators: strncpy()->sprintf(), reduce strlen()'sYaniv Kaul2018-09-074-33/+30
| | | | | | | | | | | | | | | | | | | | | | | xlators/cluster/afr/src/afr-common.c xlators/cluster/dht/src/dht-common.c xlators/cluster/dht/src/dht-rebalance.c xlators/cluster/stripe/src/stripe-helpers.c strncpy may not be very efficient for short strings copied into a large buffer: If the length of src is less than n, strncpy() writes additional null bytes to dest to ensure that a total of n bytes are written. Instead, use snprintf(). Also: - save the result of strlen() and re-use it when possible. - move from strlen to SLEN (sizeof() ) for const strings. Compile-tested only! Change-Id: Icdf79dd3d9f9ff120e4720ff2b8bd016df575c38 updates: bz#1193929 Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
* multiple: Fix python2 to python3 compatKotresh HR2018-09-073-4/+4
| | | | | | Change-Id: Iac241166d7a35dc7cc6cf07850f9f1bce38fe207 Updates: #411 Signed-off-by: Kotresh HR <khiremat@redhat.com>
* libglusterfs/src/dict.c: remove memcpy(), gf_asprintf -> asprintfYaniv Kaul2018-09-062-220/+68
| | | | | | | | | | | | | | | | | | | | | 1. We don't need a temp. string to strotol(), we can use data->data. 2. We don't need the gf_asprintf, when a regular asprintf would do. 3. No need to CALLOC the data, can MALLOC, since we are going to populate most of the fields anyway. 4. Removed is_const member - it was not used. 5. Turned is_static to a gf_boolean_t 6. Turned another get_mem0 to get_mem (in dict_set_lk() func) 7. Remove dict_set_ptr - was exactly the same as dict_set_static_ptr (apart from the 'static' bit) and no one was using it. 8. Removed is_static from the dictionary object. Did not seem to be very useful. Compile-tested only! updates: bz#1193929 Signed-off-by: Yaniv Kaul <ykaul@redhat.com> Change-Id: I7b1f6320867b154be4a94033819407f7e51cf38c
* xlators/mgmt/glusterd/src/glusterd-log-ops.c : reduce size or re-scope ↵Yaniv Kaul2018-09-061-1/+1
| | | | | | | | | | | | | | | message variable The the error and/or message variable was either: - Reduced in size - from 2048 bytes to 64 bytes, for example. or - Changed in scope - defined in a smaller scope. Compile-tested only! Change-Id: Ib8617db6c3646954c0225d12b904d668bf0f7046 updates: bz#1193929 Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
* posix: disable open/read/write on special filesAmar Tumballi2018-09-061-0/+33
| | | | | | | | | | | | | | | | | | | In the file system, the responsibility w.r.to the block and char device files is related to only support for 'creating' them (using mknod(2)). Once the device files are created, the read/write syscalls for the specific devices are handled by the device driver registered for the specific major number, and depending on the minor number, it knows where to read from. Hence, we are at risk of reading contents from devices which are handled by the host kernel on server nodes. By disabling open/read/write on the device file, we would be safe with the bypass one can achieve from client side (using gfapi) Fixes: bz#1625096 Change-Id: I48c776b0af1cbd2a5240862826d3d8918601e47f Signed-off-by: Amar Tumballi <amarts@redhat.com>
* cluster/dht: Initialise pointers to nullN Balachandran2018-09-061-2/+2
| | | | | | | | | | | | | | | | Use calloc in dht_layouts_init so to as to prevent dht_init from attempting to free invalid memory in case of failure. There are other ways to do this (set first failure to null and break there when cleaning up) but I prefer having all pointers initialized to null. This is a one time operation so it should not be too expensive. Change-Id: Ie22246047448f1cae971d48fa5aaf2efcaeb42c0 fixes: bz#1625643 Signed-off-by: N Balachandran <nbalacha@redhat.com>
* cli/src/cli-cmd-parser.c: Move to GF_MALLOC() instead of GF_CALLOC() when ↵Yaniv Kaul2018-09-051-4/+5
| | | | | | | | | | | | | | | | | | | | | | | | | possible It doesn't make sense to calloc (allocate and clear) memory when the code right away fills that memory with data. It may be optimized by the compiler, or have a microscopic performance improvement. In some cases, also changed allocation size to be sizeof some struct or type instead of a pointer - easier to read. In some cases, removed redundant strlen() calls by saving the result into a variable. 1. Only done for the straightforward cases. There's room for improvement. 2. Please review carefully, especially for string allocation, with the terminating NULL string. Only compile-tested! updates: bz#1193929 Signed-off-by: Yaniv Kaul <ykaul@redhat.com> Change-Id: Ic1531afbf614823628cc8d70bae782d4db7ec033
* features/uss: Use xxh64 to generate gfid instead of md5sumRaghavendra Manjunath2018-09-057-14/+219
| | | | | | | | | | | | | | | | * This is to ensure FIPS support * Also changed the signature of svs_uuid_generate to get xlator argument * Added xxh64 wrapper functions in common-utils to generate gfid using xxh64 - Those wrapper functions can be used by other xlators as well to generate gfids using xxh64. But as of now snapview-server is going to be the only consumer. Change-Id: Ide66573125dd74122430cccc4c4dc2a376d642a2 Updates: #230 Signed-off-by: Raghavendra Manjunath <raghavendra@redhat.com> Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
* io-stats: dump io-stats info in /var/run/glusterAmar Tumballi2018-09-052-15/+31
| | | | | | | | | | | | | | | | | | | | | | | | | | | It wouldn't make sense to allow iostats file to be written in *any* directory. While the formating makes sure we try to append io-stats-name for the file, so overwriting existing file is slim, but in any case it makes sense to restrict dumping to one directory. Below are the sample commands, and files created for the corresponding values: $ setfattr -n trusted.io-stats-dump -v file-for-dump $M0 In this case, the file would be in /var/run/gluster/file-for-dump $ setfattr -n trusted.io-stats-dump -v /dir1/dir2/file-for-dump $M0 In this case, then the dump file is in /var/run/gluster/dir1-dir2-file-for-dump Note that the value passed for this virtual xattr would be treated as a file, and even if the value has '/' in it, it would be changed to '-' for sanity. Fixes: bz#1625106 Change-Id: Id9ae6a40a190b8937c51662e6e1c2a0f6c86a0e0 Signed-off-by: Amar Tumballi <amarts@redhat.com>
* protocol: don't use allocaAmar Tumballi2018-09-052-103/+58
| | | | | | | | | | | | current implementation of alloca can cause issues when strings larger than the allocated buffer is passed to the xdr. Hence it makes sense to allow XDR decode functions to deal with memory allocations, which we can free later. Fixes: bz#1625097 Change-Id: I3a05553f5702de9575c244649ca0e5ac9abaac94 Signed-off-by: Amar Tumballi <amarts@redhat.com>
* server-protocol: don't allow '../' path in 'name'Amar Tumballi2018-09-052-0/+18
| | | | | | | | | | | | | This will prevent any arbitrary file creation through glusterfs by modifying the client bits. Also check for the similar flaw inside posix too, so we prevent any changes in layers in-between. Fixes: bz#1625095 Signed-off-by: Amar Tumballi <amarts@redhat.com> Change-Id: Id9fe0ef6e86459e8ed85ab947d977f058c5ae06e
* dict: handle negative key/value length while unserializeAmar Tumballi2018-09-051-0/+7
| | | | | | Fixes: bz#1625089 Change-Id: Ie56df0da46c242846a1ba51ccb9e011af118b119 Signed-off-by: Amar Tumballi <amarts@redhat.com>
* afr: thin-arbiter read txn changesRavishankar N2018-09-055-19/+320
| | | | | | | | | | | | | | | | If both data bricks are up, read subvol will be based on read_subvols. If only one data brick is up: - First qeury the data-brick that is up. If it blames the other brick, allow the reads. - If if doesn't, query the TA to obtain the source of truth. TODO: See if in-memory state can be maintained for read txns (BZ 1624358). updates: bz#1579788 Change-Id: I61eec35592af3a1aaf9f90846d9a358b2e4b2fcc Signed-off-by: Ravishankar N <ravishankar@redhat.com>