July 2019 - Devel - NFS Ganesha List Archives

FW: The Fall Bakeathon in Westford, Ma
by Frank Filz 08 Jul '19

08 Jul '19

-----Original Message----- From: linux-nfs-owner(a)vger.kernel.org [mailto:linux-nfs-owner@vger.kernel.org] On Behalf Of Steve Dickson Sent: Monday, July 8, 2019 11:38 AM To: NFSv4 <nfsv4(a)ietf.org>; Linux NFS Mailing list <linux-nfs(a)vger.kernel.org>; Gerard Thornley <gerard.thornley(a)hitachivantara.com>; Ben Woodard <woodard(a)redhat.com>; Karen Precourt <kprecour(a)redhat.com>; George Birner <gbirner(a)redhat.com> Subject: The Fall Bakeathon in Westford, Ma Hello all.... Happy Summer!!! Red Hat is pleased to announce the Fall NFS Bake-a-ton will be again hosted by Red Hat in Westford, MA office. I've very hopeful a couple new server vendors will be attending... and hopefully to do some new RoCE testing as well as the usual NFSoRDMA and v4.x testing. The Date: *the 2ed full week in Oct (14th - 18th)* Details and registration info: http://www.nfsv4bat.org/Events/2019/Oct/BAT/index.html I'm using the same hotels and the prices are the same or lower. The block reservations for the Residence Inn and Hampton Inn end on *Wed Sept 14th* * Residence Inn Marriott - closest; w/ shuttle; full kitchen; free breakfast * Westford Regency Inn - w/bar ; w/shuttle * Hampton Inn - free breakfast When registering (i.e. replying to this email) please give me * Company Name * How engineers are coming * T-shirt sizes * Hotel you will be staying in Any questions... Feel free to reach out... I hope to see you in October!!! steved.

1 0

Change in ...nfs-ganesha[next]: MDCACHE - Cancel dirmap LRU when destroying export
by Daniel Gryniewicz (GerritHub) 08 Jul '19

08 Jul '19

Daniel Gryniewicz has uploaded this change for review. ( https://review.gerrithub.io/c/ffilz/nfs-ganesha/+/460866 ) Change subject: MDCACHE - Cancel dirmap LRU when destroying export ...................................................................... MDCACHE - Cancel dirmap LRU when destroying export The dirmap LRU thread accesses the export, but when the export is removed, this causes a use-after-free. Cancel the dirmap LRU when it's export is removed. Change-Id: Idd001de7558234c108d9023e68609acce7d9b3bf Signed-off-by: Daniel Gryniewicz <dang(a)redhat.com> --- M src/FSAL/Stackable_FSALs/FSAL_MDCACHE/mdcache_export.c M src/FSAL/Stackable_FSALs/FSAL_MDCACHE/mdcache_int.h M src/FSAL/Stackable_FSALs/FSAL_MDCACHE/mdcache_lru.c M src/FSAL/Stackable_FSALs/FSAL_MDCACHE/mdcache_lru.h 4 files changed, 47 insertions(+), 5 deletions(-) git pull ssh://review.gerrithub.io:29418/ffilz/nfs-ganesha refs/changes/66/460866/1 -- To view, visit https://review.gerrithub.io/c/ffilz/nfs-ganesha/+/460866 To unsubscribe, or for help writing mail filters, visit https://review.gerrithub.io/settings Gerrit-Project: ffilz/nfs-ganesha Gerrit-Branch: next Gerrit-Change-Id: Idd001de7558234c108d9023e68609acce7d9b3bf Gerrit-Change-Number: 460866 Gerrit-PatchSet: 1 Gerrit-Owner: Daniel Gryniewicz <dang(a)redhat.com> Gerrit-MessageType: newchange

1 0

Change in ...nfs-ganesha[next]: Fix leak when cancelling thread fridge
by Daniel Gryniewicz (GerritHub) 08 Jul '19

08 Jul '19

Daniel Gryniewicz has uploaded this change for review. ( https://review.gerrithub.io/c/ffilz/nfs-ganesha/+/460865 ) Change subject: Fix leak when cancelling thread fridge ...................................................................... Fix leak when cancelling thread fridge Change-Id: Iba77ac2e7e1612fcba5052d72b121c0e0b4793f4 Signed-off-by: Daniel Gryniewicz <dang(a)redhat.com> --- M src/support/fridgethr.c 1 file changed, 1 insertion(+), 0 deletions(-) git pull ssh://review.gerrithub.io:29418/ffilz/nfs-ganesha refs/changes/65/460865/1 -- To view, visit https://review.gerrithub.io/c/ffilz/nfs-ganesha/+/460865 To unsubscribe, or for help writing mail filters, visit https://review.gerrithub.io/settings Gerrit-Project: ffilz/nfs-ganesha Gerrit-Branch: next Gerrit-Change-Id: Iba77ac2e7e1612fcba5052d72b121c0e0b4793f4 Gerrit-Change-Number: 460865 Gerrit-PatchSet: 1 Gerrit-Owner: Daniel Gryniewicz <dang(a)redhat.com> Gerrit-MessageType: newchange

1 0

Change in ...nfs-ganesha[next]: Do not show queue latencies in ganesha_stats
by Sachin Punadikar (GerritHub) 08 Jul '19

08 Jul '19

Sachin Punadikar has uploaded this change for review. ( https://review.gerrithub.io/c/ffilz/nfs-ganesha/+/460834 ) Change subject: Do not show queue latencies in ganesha_stats ...................................................................... Do not show queue latencies in ganesha_stats With latest Ganesha, there are no RPC queues maintained. In that context there is no meaning of fetching and displaying queue latencies. Change-Id: If91ae3454180d8562b778973d0675f2c0303b49d Signed-off-by: Sachin Punadikar <psachin(a)in.ibm.com> --- M src/include/server_stats_private.h M src/scripts/ganeshactl/Ganesha/glib_dbus_stats.py M src/scripts/ganeshactl/Ganesha/io_stats.py M src/support/server_stats.c 4 files changed, 22 insertions(+), 55 deletions(-) git pull ssh://review.gerrithub.io:29418/ffilz/nfs-ganesha refs/changes/34/460834/1 -- To view, visit https://review.gerrithub.io/c/ffilz/nfs-ganesha/+/460834 To unsubscribe, or for help writing mail filters, visit https://review.gerrithub.io/settings Gerrit-Project: ffilz/nfs-ganesha Gerrit-Branch: next Gerrit-Change-Id: If91ae3454180d8562b778973d0675f2c0303b49d Gerrit-Change-Number: 460834 Gerrit-PatchSet: 1 Gerrit-Owner: Sachin Punadikar <psachin(a)in.ibm.com> Gerrit-MessageType: newchange

1 0

Announce Push of V2.8.1.1
by Frank Filz 05 Jul '19

05 Jul '19

Branch next Tag:V2.8.1.1 Release Highlights * python: python37 inconsistent indentation style bytecompile errors * Release memory allocated to callback client * Fix displaying all protocol proc counters Signed-off-by: Frank S. Filz <ffilzlnx(a)mindspring.com> Contents: 5665fd8 Frank S. Filz V2.8.1.1 1906d78 Malahal Naineni Fix displaying all protocol proc counters 1b4cc2a Sachin Punadikar Release memory allocated to callback client 441d2e1 Kaleb S. KEITHLEY python: python37 inconsistent indentation style bytecompile errors

1 0

Change in ...nfs-ganesha[next]: Release memory allocated to callback client
by Sachin Punadikar (GerritHub) 05 Jul '19

05 Jul '19

Sachin Punadikar has uploaded this change for review. ( https://review.gerrithub.io/c/ffilz/nfs-ganesha/+/460591 ) Change subject: Release memory allocated to callback client ...................................................................... Release memory allocated to callback client For every async request being sent for NLM protocol, CLIENT structure gets allocated for the callback client. This needs to be freed when request has been sent Change-Id: Idecb63c5cbd1368df86a3381055c6d795fd1c9b5 Signed-off-by: Sachin Punadikar <psachin(a)in.ibm.com> --- M src/SAL/nlm_owner.c 1 file changed, 5 insertions(+), 0 deletions(-) git pull ssh://review.gerrithub.io:29418/ffilz/nfs-ganesha refs/changes/91/460591/1 -- To view, visit https://review.gerrithub.io/c/ffilz/nfs-ganesha/+/460591 To unsubscribe, or for help writing mail filters, visit https://review.gerrithub.io/settings Gerrit-Project: ffilz/nfs-ganesha Gerrit-Branch: next Gerrit-Change-Id: Idecb63c5cbd1368df86a3381055c6d795fd1c9b5 Gerrit-Change-Number: 460591 Gerrit-PatchSet: 1 Gerrit-Owner: Sachin Punadikar <psachin(a)in.ibm.com> Gerrit-MessageType: newchange

1 0

Change in ...nfs-ganesha[next]: Fix displaying all protocol proc counters
by Malahal (GerritHub) 03 Jul '19

03 Jul '19

Malahal has uploaded this change for review. ( https://review.gerrithub.io/c/ffilz/nfs-ganesha/+/460261 Change subject: Fix displaying all protocol proc counters ...................................................................... Fix displaying all protocol proc counters The last one is not sent over the dbus currently. Fix it by sending the last entry. Also, fix the wrong use of the limit for NLM operations! Change-Id: I3cc93366e7d9e39a1b82193faea3ff596d3cf9a1 Signed-off-by: Malahal Naineni <malahal(a)us.ibm.com> --- M src/support/server_stats.c 1 file changed, 4 insertions(+), 4 deletions(-) git pull ssh://review.gerrithub.io:29418/ffilz/nfs-ganesha refs/changes/61/460261/1 -- To view, visit https://review.gerrithub.io/c/ffilz/nfs-ganesha/+/460261 To unsubscribe, or for help writing mail filters, visit https://review.gerrithub.io/settings Gerrit-Project: ffilz/nfs-ganesha Gerrit-Branch: next Gerrit-Change-Id: I3cc93366e7d9e39a1b82193faea3ff596d3cf9a1 Gerrit-Change-Number: 460261 Gerrit-PatchSet: 1 Gerrit-Owner: Malahal <malahal(a)gmail.com> Gerrit-MessageType: newchange

1 0

Announce Push of V2.8.1
by Frank Filz 02 Jul '19

02 Jul '19

Branch next Tag:V2.8.1 Release Highlights * 2 readdir fixes Signed-off-by: Frank S. Filz <ffilzlnx(a)mindspring.com> Contents: 22ab12e Frank S. Filz V2.8.1 445575d Daniel Gryniewicz MDCACHE - initialize free_whence to avoid suprious free c7e7d24 Daniel Gryniewicz MDCACHE - Null out chunk when restarting readdir

1 0

Change in ...nfs-ganesha[next]: MDCACHE - initialize free_whence to avoid suprious free
by Daniel Gryniewicz (GerritHub) 02 Jul '19

02 Jul '19

Daniel Gryniewicz has uploaded this change for review. ( https://review.gerrithub.io/c/ffilz/nfs-ganesha/+/460151 Change subject: MDCACHE - initialize free_whence to avoid suprious free ...................................................................... MDCACHE - initialize free_whence to avoid suprious free Change-Id: Ia2f1e4b1c4bcde914276c54412ae7ad6cf75d1d1 Signed-off-by: Daniel Gryniewicz <dang(a)redhat.com> --- M src/FSAL/Stackable_FSALs/FSAL_MDCACHE/mdcache_helpers.c 1 file changed, 1 insertion(+), 1 deletion(-) git pull ssh://review.gerrithub.io:29418/ffilz/nfs-ganesha refs/changes/51/460151/1 -- To view, visit https://review.gerrithub.io/c/ffilz/nfs-ganesha/+/460151 To unsubscribe, or for help writing mail filters, visit https://review.gerrithub.io/settings Gerrit-Project: ffilz/nfs-ganesha Gerrit-Branch: next Gerrit-Change-Id: Ia2f1e4b1c4bcde914276c54412ae7ad6cf75d1d1 Gerrit-Change-Number: 460151 Gerrit-PatchSet: 1 Gerrit-Owner: Daniel Gryniewicz <dang(a)redhat.com> Gerrit-MessageType: newchange

1 0

2.7.4 core in mdcache (includes the patches)
by Rungta, Vandana 02 Jul '19

02 Jul '19

Daniel, Core in mdcache when mdcache_lru_unref_chunk. Ref count is 0. I started these tests 10 days ago with all the mdcache patches including the one moving the unref_chunk in readdir_chunked to inside the lock. The test continues to readdir over 5 million entries while writing/reading/deleting content. I have 2 test setups – both using a Windows client that hit this core. Let me know if there is additional information you need from the cores. Thanks, Vandana [Thread debugging using libthread_db enabled] Using host libthread_db library "/lib64/libthread_db.so.1". Core was generated by `bin/ganesha.nfsd -f etc/ganesha/ganesha.conf -p var/run/ganesha.pid -F'. Program terminated with signal 11, Segmentation fault. #0 0x00007fbd14d39c40 in pthread_mutex_lock () from /lib64/libpthread.so.0 Missing separate debuginfos, use: debuginfo-install sgw-nfs-ganesha-2.0.93.0-1.x86_64 (gdb) bt #0 0x00007fbd14d39c40 in pthread_mutex_lock () from /lib64/libpthread.so.0 #1 0x000000000052af4a in _mdcache_lru_unref_chunk (chunk=0x3836d1a0, func=0x598a00 <__func__.23678> "mdcache_readdir_chunked", line=3133) at /src/src/FSAL/Stackable_FSALs/FSAL_MDCACHE/mdcache_lru.c:2058 #2 0x0000000000542d84 in mdcache_readdir_chunked (directory=0x664fa280, whence=37935356, dir_state=0x7fbd0a3dfaf0, cb=0x4323ed <populate_dirent>, attrmask=0, eod_met=0x7fbd0a3dffeb) at /src/src/FSAL/Stackable_FSALs/FSAL_MDCACHE/mdcache_helpers.c:3133 #3 0x000000000053054c in mdcache_readdir (dir_hdl=0x664fa2b8, whence=0x7fbd0a3dfad0, dir_state=0x7fbd0a3dfaf0, cb=0x4323ed <populate_dirent>, attrmask=0, eod_met=0x7fbd0a3dffeb) at /src/src/FSAL/Stackable_FSALs/FSAL_MDCACHE/mdcache_handle.c:559 #4 0x0000000000432d14 in fsal_readdir (directory=0x664fa2b8, cookie=37935356, nbfound=0x7fbd0a3dffec, eod_met=0x7fbd0a3dffeb, attrmask=0, cb=0x491d35 <nfs3_readdir_callback>, opaque=0x7fbd0a3dffa0) at /src/src/FSAL/fsal_helper.c:1164 #5 0x0000000000491b1d in nfs3_readdir (arg=0xa8fc5d48, req=0xa8fc5640, res=0xa32fbb30) at /src/src/Protocols/NFS/nfs3_readdir.c:289 #6 0x0000000000457c16 in nfs_rpc_process_request (reqdata=0xa8fc5640) at /src/src/MainNFSD/nfs_worker_thread.c:1328 #7 0x00000000004583d5 in nfs_rpc_valid_NFS (req=0xa8fc5640) at /src/src/MainNFSD/nfs_worker_thread.c:1548 #8 0x00007fbd15d14034 in svc_vc_decode (req=0xa8fc5640) at /src/src/libntirpc/src/svc_vc.c:829 #9 0x000000000044adc5 in nfs_rpc_decode_request (xprt=0x7ae54c90, xdrs=0x36434390) at /src/src/MainNFSD/nfs_rpc_dispatcher_thread.c:1345 #10 0x00007fbd15d13f45 in svc_vc_recv (xprt=0x7ae54c90) at /src/src/libntirpc/src/svc_vc.c:802 #11 0x00007fbd15d10689 in svc_rqst_xprt_task (wpe=0x7ae54ea8) at /src/src/libntirpc/src/svc_rqst.c:769 #12 0x00007fbd15d10ae6 in svc_rqst_epoll_events (sr_rec=0x1acd4d0, n_events=1) at /src/src/libntirpc/src/svc_rqst.c:941 #13 0x00007fbd15d10d7b in svc_rqst_epoll_loop (sr_rec=0x1acd4d0) at /src/src/libntirpc/src/svc_rqst.c:1014 #14 0x00007fbd15d10e2e in svc_rqst_run_task (wpe=0x1acd4d0) at /src/src/libntirpc/src/svc_rqst.c:1050 #15 0x00007fbd15d197f6 in work_pool_thread (arg=0x75491e70) at /src/src/libntirpc/src/work_pool.c:181 #16 0x00007fbd14d37de5 in start_thread () from /lib64/libpthread.so.0 #17 0x00007fbd1463ef1d in clone () from /lib64/libc.so.6 (gdb) select-frame 1 (gdb) info locals rc = 32701 refcnt = 171833376 lane = 268435456 qlane = 0xe007e24e0 __func__ = "_mdcache_lru_unref_chunk" (gdb) info args chunk = 0x3836d1a0 func = 0x598a00 <__func__.23678> "mdcache_readdir_chunked" line = 3133 (gdb) print *chunk $1 = {chunks = {next = 0xed43cd15, prev = 0xa386010002000000}, dirents = {next = 0x300000003000000, prev = 0x1c00000001000000}, parent = 0x7000000aaaaaaaa, chunk_lru = {q = {next = 0x6e776f6e6b6e75, prev = 0xfefffffffeffffff}, qid = LRU_ENTRY_NONE, refcnt = 0, flags = 0, lane = 268435456, cf = 16777283}, reload_ck = 1099511627778, next_ck = 7236828793636126720, num_entries = 841903471} (gdb) select-frame 2 (gdb) info locals status = {major = ERR_FSAL_NOENT, minor = 0} cb_result = DIR_CONTINUE entry = 0x0 attrs = {request_mask = 0, valid_mask = 1433550, supported = 1433582, type = REGULAR_FILE, filesize = 51682, fsid = {major = 0, minor = 0}, acl = 0x0, fileid = 37935372, mode = 493, numlinks = 1, owner = 4294967294, group = 4294967294, rawdev = {major = 0, minor = 0}, atime = { tv_sec = 1561811134795000, tv_nsec = 0}, creation = {tv_sec = 0, tv_nsec = 0}, ctime = {tv_sec = 1561811134795000, tv_nsec = 0}, mtime = { tv_sec = 1561795757, tv_nsec = 842000000}, chgtime = {tv_sec = 1561811134795000, tv_nsec = 0}, spaceused = 51682, change = 1561811134795000000, generation = 0, expire_time_attr = 60, fs_locations = 0x0, sec_label = {slai_lfs = {lfs_lfs = 0, lfs_pi = 0}, slai_data = {slai_data_len = 0, slai_data_val = 0x0}}} dirent = 0xa57f3030 has_write = true set_first_ck = false next_ck = 37935375 look_ck = 37935377 chunk = 0x3836d1a0 first_pass = true eod = false reload_chunk = false __func__ = "mdcache_readdir_chunked" __PRETTY_FUNCTION__ = "mdcache_readdir_chunked" (gdb) info args directory = 0x664fa280 whence = 37935356 dir_state = 0x7fbd0a3dfaf0 cb = 0x4323ed <populate_dirent> attrmask = 0 eod_met = 0x7fbd0a3dffeb (gdb)

2 5