5.4.1 open issues
From Net-SNMP Wiki
We try to collect significant open issues for the 5.4.1 release cycle here. Please add your contributions/comments and fix as much as you can!
Open issues for 5.4.1
|1||ipAddressTable crash with double free||Bug 1733344||Crashes on FC6 with double free. Backtrace available.||rstory|
|4||set O_NONBLOCK on Linux!||Bug 1718296||recvfrom can block if we don't set O_NONBLOCK on Linux||?|
|5||HP-UX 11.23/IA64 build failure||Bug 1742278||Build fails in snmplib/system.c. Preprocessor output available.|| tanders/magfr?
|6||hrStorageSize overflow||Bug 1743171||hrStorageSize is negative for large filesystems, although the MIB only allows values >=0.||?|
|7||error on subcontainer 'ia_addr' insert (-1)||Bug 1693039, Patch 1712645||Provide a more meaningful log message for duplicate IP addresses instead of spamming the log continuously.||rstory?|
|8||ifTable transient interface handling||Bug 1238981, Patch 1513191|| offer an (per-interface) config option how transient interfaces are handled in ifTable (same or different). A List FAQ.
This feels like a new feature, and not appropriate for 5.4.x?
|9||Suppress error messages on SIGHUP||?|| Remove the likely harmless, but annoying error messages on SIGHUP. A List FAQ
Handled for "extend" and "monitor" directives. What others?
|10||disman/event crash||Bug 1556259|| Reported crashes in mteTrigger.c
This is a race condition between (typically high-volume) request handling and trap generation. Not realistic to address for 5.4.1?
|11||Solaris perl<->host issues (due to largefiles)||Bug 1486366|| Solaris system-supplied Perl is compiled with largefile support. The very same options break the build of the host module, though.
At least detected (and embedded perl disabled) since 5.4.pre3.
|12||64bit nlist||Bugs 148532, 1484814, 1388562||nlist fails on 64bit systems (AIX, IRIX, Tru64). We need 64bit capable auto_nlist replacement routines. Every system has its own variant and changes are needed all over the net-snmp code. Stop-gap fix is specifying --without-kmem-usage which is now possible since 5.4.rc2.||tanders|
|"exec/sh /bin/true" hangs the agent||Bug 1745113|| "exec truth /bin/true" causes the agent to hang while eating 100% cpu
Fixed in SVN revision 16594
|Perl linking||-||libperl is still incorrectly linked to the applications rather than libnetsnmpagent/libnetsnmptrapd. 5.4.1.rc1 allows to manually configure --enable-as-needed (default to off in 5.4, on in trunk).||tanders|
|OSX embedded perl broken||Bug 1377478||Embedded perl (and sometimes even the perl modules) seem to fail silently (and for unknown reasons) on OSX for some people, but not all. Should now be fixed in 5.4.1.rc1 when using proper libperl linking (--enable-as-needed).||tanders|
|AIX embedded perl build broken||Bug 1600099||Embedded perl fails to build on AIX, even with properly built shared libraries. It's a xlc/cc_r compiler issue. Now documented in README.aix.||tanders|
|library layering||Bug 1565496 (OSX), Patches 1398873 1493522||Mostly fixed for 5.4.1.pre2. Full story here.||tanders|
|Python build||-||There's a Debian patch (#38) to allow for in-tree building of the Python modules.||tanders|
|snmptrap/trapsess send to port 161||Patch 1620424||snmptrap and trapsess send to port 161 instead of port 162 by default. Patch from magfr available.||magfr|
|Linux CPU stats||Bug 1600261|| Broken for Montavista Linux on XScale/ARM and possibly others. See bug description. Regression over 5.3.x.
Appears to be very hardware-specific. Fixed in 5.4.x SVN (r15839).
|fix perl compiler warnings||-||There are a number of nasty compiler warnings for the perl modules that needs to be dealt with. List provided to hardaker. (http://pastebin.ca/196857) Dave has flagged three issues as potentially important after review (see list). Full fixes for SNMP.pm are in 5.4.x CVS.||tanders|
|OSX CPU stats||Patch 1600522|| Neither cpu_sysctl (preferrable?) nor cpu_nlist compile on Darwin. rc1/rc2 ship cpu_null for Darwin. To be investigated! At least it doesn't seem to be a regression over 5.3.x, based on rstory's snmpwalk output for darwin7/8.
A new cpu_mach.c has been supplied as a patch after rc3!
Too late for 5.4. Now in 5.4.x SVN.
|libtool 1.5.24?||-||5.4.1.pre3 ships with libtool 1.5.22. Is it worth upgrading to 1.5.24 (as in trunk)? Upgraded to 1.5.24 before 5.4.1.rc4.||tanders|
Documentation and testing work for 5.4.1
|manual pages||quite a few manual pages currently ship with explicit XXX - to be filled in for 5.4.1 final.||Volunteers?|
|missing tests|| We're currently missing tests for at least:
disman/event monitoring, snmpvacm extensions, snmptranslate, snmpusm DH stuff, snmptrapd walk, ...