From 2015-06-14 to 2015-07-13
- 08:21 PM Bug #6053: update usr/src/uts/common/sys/feature_tests.h
- 10:15 AM Bug #6053 (In Progress): update usr/src/uts/common/sys/feature_tests.h
- need update usr/src/uts/common/sys/feature_tests.h for better portability of userland components.
- 08:06 PM Feature #5942 (Closed): integrate dmake
- "git commit 10d63b7db37a83b39c7f511cf9426c9d03ea0760":https://github.com/illumos/illumos-gate/commit/10d63b7db37a83b3...
- 03:49 PM Bug #2921 (In Progress): Many manpages asking for tbl do not need it
- 03:43 PM Bug #1438 (In Progress): ftw(3C) should mention `quit' member of struct FTW and possible values
- 11:09 AM Bug #6052 (Closed): decouple lzc_create() from the implementation details
- At the moment @type@ parameter of @lzc_create()@ is of @dmu_objset_type_t@ type.
That exposes an implementation deta...
- 10:58 AM Feature #6051: lzc_receive: allow the caller to read the begin record
- Oops, a symbol was "eaten" by the markup.
The first line should read:
> Currently lzc_receive() requires that its s...
- 10:56 AM Feature #6051 (Closed): lzc_receive: allow the caller to read the begin record
- Currently @lzc_receive()@ requires that its @snapname@ argument is a snapshot name (contains '@').
@zfs receive@ all...
- 06:29 PM Bug #6050 (Closed): kstat_create(9f) should document KSTAT_TYPE_RAW and KSTAT_TYPE_TIMER
- The kstat_create(9f) man page should document both KSTAT_TYPE_RAW and KSTAT_TYPE_TIMER kstat types.
- 04:03 PM Bug #5910: libnisdb won't build with modern GCC
- I just noticed that `CFLAGS += $(CCVERBOSE)' is missing from the Makefile . I'll try adding that to see what develop...
- 02:45 PM Feature #3188: ignore the comment in limits.h and define NAME_MAX
- Surely it would be (MAXNAMELEN-1) which would be 255. NAME_MAX is not supposed to include a terminating NUL.
- 02:41 PM Bug #1926: libresolv evades compiler warnings
- I'll take care of this one.
- 03:26 PM Feature #5265: rearchitect and replace interrupt distribution
- Adding this for reference: https://hg.openindiana.org/projects/illumos/intrd/summary
- 05:11 AM Bug #6048 (Closed): RLIM_INFINITY (et al) should be unsigned for _LP64
- building gcc with gcc for 64-bits (-m64) identifies an issue in /usr/include/sys/resource.h
with the definition of R...
- 08:30 PM Feature #6047 (Closed): SPARC boot should support feature@embedded_data
- another missing feature
- 08:29 PM Feature #6046 (Closed): SPARC boot should support com.delphix:hole_birth
- missing feature
- 07:44 PM Feature #5788: Want support for GCC's stack protector in libc
- I haven't seen it either. It must be disabled on Solaris/illumos, probably due to the issues I mentioned above.
- 07:40 PM Feature #5788: Want support for GCC's stack protector in libc
- The "on by default in newer gcc" from a similar bug had me concerned. I see no libssp linkages in omnios-build built...
- 07:37 PM Feature #5788: Want support for GCC's stack protector in libc
- It kind of does cause problems. If you pass -fstack-protector when compiling libssp gets linked in automatically, bu...
- 06:35 PM Feature #5788: Want support for GCC's stack protector in libc
- Interesting. I've started using gcc51 in OmniOS's bloody for building non-ON bits. Does libc's lack of support here...
- 06:59 PM Bug #687 (Closed): rpcgen should not generate absolute #includes
- "git commit 77eb9dd4fa1a2a4b00af3d1973a84b60fbf40cdf":https://github.com/illumos/illumos-gate/commit/77eb9dd4fa1a2a4b...
- 04:59 PM Bug #5172 (Closed): tftpd service is not registered by default and its manual is giving misleading directions
- "git commit 29a47e6ee6365b8ebcd68a3d59c0e1ef32f17ceb":https://github.com/illumos/illumos-gate/commit/29a47e6ee6365b8e...
- 08:47 AM Bug #6045: zfs_mount(MS_REMOUNT) is unsafe
- Clarification: @z_teardown_lock@ lock should be taken in the writer / exclusive mode.
- 08:43 AM Bug #6045 (New): zfs_mount(MS_REMOUNT) is unsafe
- zfs_mount(MS_REMOUNT) calls @zfs_unregister_callbacks()@ + @zfs_register_callbacks()@ without any synchronization wit...
- 08:25 AM Bug #6044 (Closed): SPARC zfs reader is using wrong size for objset_phys
- current get-objset is using 512B buffer length to read from bp, but objset_phys is 2048 bytes.
- 10:31 PM Feature #6043 (New): mdb: pipe shell output back into mdb ("! ... !")
- MDB lets you pipe the output to a subshell using `!`, as in:...
- 10:19 PM Feature #6042 (New): mdb buffers pipelines too much
- In "mdb -k", I had a pointer to a list of 48M taskq entries. When I do this:...
- 02:46 PM Bug #6033 (Closed): arc_adjust() should search MFU lists for oldest buffer when adjusting MFU size
- "git commit 31c46cf23cd1cf4d66390a983dc5072d7d299ba2":https://github.com/illumos/illumos-gate/commit/31c46cf23cd1cf4d...
- 03:29 PM Bug #4051: Don't free mdi_pathinfo_t in mptsas when device(s) are retired.
- Change summary, original was "crash in vhci_mpapi_sync_lu_oid_list() on ioctl, possible SAS problem".
Changed it t...
- 09:29 AM Feature #6041 (Closed): SPARC boot should support LZ4
- As lz4 is now default, SPARC zfs bootblock needs update to support LZ4.
- 06:16 PM Bug #6039 (Closed): installgrub MBR update response check should use locale specific yesexpr
- As installgrub messages are localised, the response for MBR update question should be checked against locale yesexpr.
- 12:18 AM Bug #4085 (Closed): mpathadm list lu cause panic
- 12:18 AM Bug #4085 (Feedback): mpathadm list lu cause panic
- Closing as a duplicate of #4051.
- 12:15 AM Bug #4051: Don't free mdi_pathinfo_t in mptsas when device(s) are retired.
- Here's the analysis from #4085:
This has the same symptoms as an issue with retired devices with multiple paths we...
- 10:14 PM Bug #6037 (In Progress): zfs(1M) needs to handle unknown uid/gid in context of allow/unallow more gracefully
- Currently, if a uid/gid is unknown to the system, zfs(1M) would say nothing on the unknown user/group. For instance,...
- 05:31 PM Bug #6036 (Closed): ufsrestore segfaults in lookupparent
- One of the zfs-tests (functional/inuse/inuse_003_pos.ksh) uses ufsrestore which happens to segfault. I don't think i...
- 12:39 AM Bug #6033 (In Progress): arc_adjust() should search MFU lists for oldest buffer when adjusting MFU size
- 12:33 AM Bug #6033: arc_adjust() should search MFU lists for oldest buffer when adjusting MFU size
- Alek Pinchuk wrote:
> When we're looking for the list containing oldest buffer we never actually look at the MFU lis...
- 12:21 AM Bug #6033 (Closed): arc_adjust() should search MFU lists for oldest buffer when adjusting MFU size
- When we're looking for the list containing oldest buffer we never actually look at the MFU lists even when we try to ...
- 10:33 PM Feature #6032 (In Progress): stmfadm should sanitize user-provided group names
- stmfadm does not provide proper validation for host/target group names.
At least it allows to create names with:
- 04:44 PM Bug #6031 (New): Possible memory leak in svc_dg_create_private()
- In a case the malloc() call at line 158 fails we will leak the su buffer here in svc_dg_create_private():...
- 03:50 PM Bug #6030 (New): Possible memory leak in select_server_addr()
- There is possible leak of sock at line 681 here in select_server_addr():...
- 03:31 PM Bug #6029 (New): Too big buffer is allocated in logtab_rement()
- Here in logtab_rement() there is too big buffer allocated:...
- 09:06 PM Bug #6014 (Pending RTI): rpcgen should be built with tools and using for builds
- 04:19 PM Feature #5922: Want support for building with -fstack-protector
- FWIW the changes look good to me. I might be able to help with testing if that's what's holding up the integration p...
- 04:13 PM Feature #5922: Want support for building with -fstack-protector
- Revised version: https://xylem.cooperi.net/~alex/webrev-5922-2/
- 03:08 PM Feature #5922: Want support for building with -fstack-protector
- Link to webrev that was posted to the mailing list: https://xylem.cooperi.net/~alex/webrev-5922/
- 09:42 AM Bug #6028: dd can exit prematurely upon SIGINFO/SIGUSR1
- Lauri Tirkkonen wrote:
> % dd if=/dev/zero bs=8k |ssh $(hostname) cat >/dev/null
'cat >/dev/null' should be q...
- 09:36 AM Bug #6028 (New): dd can exit prematurely upon SIGINFO/SIGUSR1
- dd, if interrupted mid-write, exits prematurely. #4494 added the signal
handlers for INFO/USR1, which is useful, but...
- 06:09 PM Bug #6014: rpcgen should be built with tools and using for builds
clean up gcc warnings too
- 02:27 PM Bug #687: rpcgen should not generate absolute #includes
- I'll look after this bug. I have a simple solution that solves the problem.
- 12:31 AM Bug #6027 (Closed): EOL zulu (XVR-4000)
- XVR-4000 was a very expensive, very rare graphics card with only a handful of them ever sold.
Removing this will p...
- 11:52 PM Bug #6026 (Closed): Zero-padded IP address strings returned by SMB
- When using MMC to list connected users, Windows has long delays updating the display for each user.
It has problems ...
- 11:49 PM Bug #6025 (Closed): Customer cannot set Share Properties using MMC
- If anything changes libshare after the SMB server starts (i.e. changes from outside SMB) then using MMC to modify sha...
- 11:48 PM Bug #6024 (Closed): ACL editor crash in Windows 2012 R2
- Windows 2012 R2 uses a new information level in the SamrQueryAliasInfo RPC call, which we did not handle.
- 11:47 PM Bug #6023 (Closed): Quota tab for an SMB share sometimes doesn't work
- There were a couple races with ZFS dataset creation and mount, where the SMB server would try too soon to setup the ....
- 11:41 PM Bug #6022 (Closed): MacOS logs error messages when trying to lookup users
- MacOS is sending LSA RPC calls: LsarGetConnectedUser when we're in workgroup mode.
Our implementation only expects ...
- 11:39 PM Bug #6021 (Closed): Windows Explorer Stops Working / Crashes when Security Tab is accessed
- When editing an acl (right-click on a directory) the Windows ACL editor tries to lookup the SIDs found in the ACL usi...
- 11:36 PM Bug #6020: panic in smbsrv netbios_first_level_name_decode
- The decoding of the "scope" field is not sufficiently careful about lengths boundary checks etc.
We didn't want to s...
- 11:31 PM Bug #6020 (Closed): panic in smbsrv netbios_first_level_name_decode
- Kernel panic seen occasionally "in the wild":...
- 02:10 PM Bug #6019 (Closed): sleep can't be infinite
- The following snippet from the sleep(1) man page is incorrect:
point number may be specified in all...
- 02:06 AM Bug #3768 (Closed): fnmatch(5) is worded poorly
- "git commit 6203546182f592d62b0fb7f4182da3d21d59031f":https://github.com/illumos/illumos-gate/commit/6203546182f592d6...
- 02:06 AM Bug #5981 (Closed): Deadlock in dmu_objset_find_dp
- "git commit 1d3f896f5469c69c1339890ec3d68e9feddb0343":https://github.com/illumos/illumos-gate/commit/1d3f896f5469c69c...
- 12:36 AM Bug #6018 (New): Network Lock Manager (NLM) fails Connectathon lock test
- Comparing the old (closed) lock manager with the current one.
With the old NLM code, this all passed (see: old-klm-...
- 08:44 PM Bug #6017 (Closed): savecore's messages about read errors are useless
- A user says...
- 07:16 PM Bug #5996 (Closed): libtopo ses module unload takes too long
- "git commit 057c620a8f2b46f09e91dfff7da884175a36b0a8":https://github.com/illumos/illumos-gate/commit/057c620a8f2b46f0...
- 06:17 PM Bug #5996: libtopo ses module unload takes too long
- I tested this using the ZFS test suite in conjunction with 5997. Without this change 5997 would make the zpool expand...
- 07:16 PM Feature #5997 (Closed): FRU field not set during pool creation and never updated
- "git commit 2750f8d5ec1b891560ac2224f6c37243d910bd1b":https://github.com/illumos/illumos-gate/commit/2750f8d5ec1b8915...
- 06:18 PM Feature #5997: FRU field not set during pool creation and never updated
- I tested this using a test machine with a 8-disk jbod attached. I created a pool on four disks, removed the other fou...
- 06:58 PM Bug #6016 (New): Some file locking scenarios does not work as expected
- h2. The lock tool
To reproduce this bug I use the attached simple interactive/scriptable locking application. To ...
- 04:16 PM Bug #6015 (In Progress): vnode: v_path reading should be protected by v_lock
- The vnode's v_path is stable only in a case the v_lock is held. The nice example of that is in the vn_copypath() fun...
- 07:25 AM Bug #5972 (Closed): kmem: fix comment typo
- "git commit 9321cd0436261240311eaac9aad60ee764e089e6":https://github.com/illumos/illumos-gate/commit/9321cd0436261240...
- 07:25 AM Bug #5973 (Closed): kmem: double-calling kmem_depot_ws_update isn't obvious
- "git commit 0c833d6473a8b5e01ef9b7bc9ac547c0a016f932":https://github.com/illumos/illumos-gate/commit/0c833d6473a8b5e0...
- 07:25 AM Bug #5974 (Closed): kmem: remove a check that's always false
- "git commit 3f2b8a52377413f4fdd822ed4aa046ba5eff81d9":https://github.com/illumos/illumos-gate/commit/3f2b8a52377413f4...
- 10:28 PM Feature #6009: Support for remote stale lock detection
- Review: https://www.illumos.org/rb/r/61/
- 03:48 PM Bug #6014 (Pending RTI): rpcgen should be built with tools and using for builds
- we are using rpcgen form build host and it is a problem if we need to use updated rpcgen for build.
- 08:56 PM Feature #4107 (In Progress): Add passwd option to read passwords from stdin
- 06:01 PM Bug #687: rpcgen should not generate absolute #includes
- i have made some change for using rpcgen from workspace(it built with tools) instead of build host.
i have found tha...
- 05:52 PM Bug #687: rpcgen should not generate absolute #includes
- I tried to reproduce this problem, but was unable to do so. It's true that idmap_xdr.c contains an absolute path to ...
- 03:47 PM Bug #3727 (Closed): british people can't spell
- "git commit 17723245e5b42ca8abb9964a36c0b051c0b237fe":https://github.com/illumos/illumos-gate/commit/17723245e5b42ca8...
- 12:14 PM Bug #5172 (In Progress): tftpd service is not registered by default and its manual is giving misleading directions
- 06:06 PM Bug #6012 (New): SMF error message when encountering I/O errors is grammatically wrong
- 04:56 PM Bug #6011: potential ZFS deadlock when ZFS writes take a page fault
- This is OmniOS 151006, revision b281e50 in the https://github.com/omniti-labs/illumos-omnios repo, so somewhat old. ...
- 04:35 PM Bug #6011 (New): potential ZFS deadlock when ZFS writes take a page fault
- An OmniOS user has a system which appeared to wedge solidly when attempting to use procfs.
Investigation finds tha...
- 03:16 PM Feature #6010 (New): zfs's refcount debugging should be generally available and more widely used
- the ZFS folks implemented a feature I have wanted for years, and hid it away in ZFS so well that I only just discover...
- 12:45 PM Bug #5219: l2arc_write_buffers() may write beyond target_sz
- I actually was wrong about which commit introduced the problem.
Indeed, commit "e14bb3258d05c1b1077e2db7cf77088924e5...
- 12:37 PM Bug #5219: l2arc_write_buffers() may write beyond target_sz
- See https://reviews.csiden.org/r/229/
- 01:11 AM Feature #6009 (In Progress): Support for remote stale lock detection
- The support for locking over NFSv2/3 is implemented using the sideband NLM protocol (with the support of the NSM prot...
- 08:50 PM Feature #5995: RPC over SMB named pipes should use AF_UNIX sockets
- "git commit bce01b59de50fe66a0267f4aa23e1d6e60d973d9":https://github.com/illumos/illumos-gate/commit/bce01b59de50fe66...
- 07:34 PM Bug #6008 (New): "you must be root to use savecore" message is not always true
- If you need more privileges than you have, savecore says "You must be root to use savecore". That's not quite true. ...
- 06:40 PM Bug #6007 (New): assertion failed: TAILQ_EMPTY(&hostp->nh_vholds_list), file: ../../common/klm/nlm_impl.c, line: 1161
- Following panic happened on the recent (commit 0afb687) debug build:...
- 01:42 PM Bug #6006: assertion failure in zdb after ZFS 'filesystem full' event
- System is oi 151a9 (update 2) with illumos rev 52e13e00ba
- 01:25 PM Bug #6006 (New): assertion failure in zdb after ZFS 'filesystem full' event
- After a 'filesystem full' event, running
zdb -vvv -d rpool | egrep '(path|size)'|grep -v Object
hit corrupt direct...
- 02:58 AM Bug #3768 (In Progress): fnmatch(5) is worded poorly
- 09:21 PM Bug #4751 (Closed): ipadm(1M) should make it clear that static v6 addrs on an interface require addrconf addrs
- "git commit b884368dd61598a2c84215e7b8620e34fba645a1":https://github.com/illumos/illumos-gate/commit/b884368dd61598a2...
- 09:21 PM Bug #2056 (Closed): update boot(1m) for right location of the GRUB menu
- "git commit 5ce9dfc844a5f2a338bcba6fcfd718a0e60956fe":https://github.com/illumos/illumos-gate/commit/5ce9dfc844a5f2a3...
- 07:17 PM Bug #6004: ld should warn about duplicate "local: *"
- Surely it should also warn for global:* in two blocks as well? I would guess the asterisk is the important bit.
- 01:52 AM Bug #6004 (New): ld should warn about duplicate "local: *"
- According to Rich Lowe, ld should warn or fail if you have "local: *" in two SYMBOL_VERSION blocks.
- 08:59 AM Feature #6005 (Closed): bootadm update-archive should create/update archive sha-1 hash
- As Illumos kernel already has simple mechanism to check module hashes at early boot (dboot_startkern.c), the hash man...
Also available in: Atom