5.7.4 candidate bugs

2018-04-13 Thread Keith Mendoza 
As I mentioned in the "Summary of meeting between NET-SNMP devs and ICEI" 
thread, here's a list of bugs that I found could be included in a 5.7.4 release.

Comment indicate change is in v5-7-patch branch, and waiting for confirmation 
from submitter
* 2810 Logical error in agent/​helpers/​table.c when --enable-read-only is used 
during configuring

* 2845 compile error with net-snmp-5.8.pre2/agent/snmp_agent.c

* 2846 Second compile error with net-snmp-5.8.pre2/agent/snmp_agent.c

* 2822 5.4 4 segfault with verison 3 and agent restart

* 2791  systemd returns 0 prCount when net-snmpd collect number of running 
processes

Comment indicate change is in v5-7-patch branch, and submitter confirms the fix 
works
* 2550 netsnmp_assert failed message

Comment indicate that the proposed patch in the patch ticket resolves the issue
* 2640 Crash in function ipAddressTable_container_load() (See patch ticket 2589)

-- 
Thanks,
Keith (pantherse)

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Net-snmp-coders mailing list
Net-snmp-coders@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders

Re: Summary of meeting between NET-SNMP devs and ICEI

2018-04-13 Thread Keith Mendoza 


On Thu, Apr 12, 2018, at 2:18 PM, Magnus Fromreide wrote:
> On Thu, Apr 12, 2018 at 12:19:35PM -0700, Keith Mendoza wrote:
> > On Thu, Apr 12, 2018, at 10:03 AM, Bill Fenner wrote:
> > > I'm sorry that I wasn't available for this meeting.  I think one important
> > > pain point is the overhead of doing releases - 5.7.3 was years ago and
> > > there are very useful fixes in the 5.7 branch; why can't we just say 
> > > "now's
> > > a good time for 5.7.4 and if we don't get it right then we can release
> > > 5.7.5 soon"?  To be fair, I've never driven a net-snmp release so I don't
> > > know what's involved.
> > 
> > I think from a project morale perspective it will be good to get a 5.7.4 
> > out. > At the very least, it'll show the world that the project is active 
> > (or active
> > again if they choose to see it that way). Is there a working list of what
> > bugs/features are planned for 5.7.4; if so, can that be finalized at this
> > point?
> 
> I think 5.7.4 is just a bugfix release.
> 5.8, of which I think the release process was initiated a while back, is
> a new feature release. I know that I have added a typed value to the perl
> handlers so that you won't have to change a global parameter in the agent to
> get hold of the numeric oid value of an oid-type variable.

I'll go ahead and look through the bugs that are in "Pending" state and compile 
a list of ones that have an indication that it has fixes ready for a  5.7.4. 
When I'm done, I'll send the list in a separate thread; hopefully, that 
provides some visibility of what a 5.4.7 might look like.

> 
> /MF
> 
> > > 
> > >   Bill
> > > 
> > > 
> > > On Thu, Apr 12, 2018 at 11:31 AM, Ian Bruene  wrote:
> > > 
> > > >
> > > > This morning we (Keith, Ian) met with an assortment of the NET-SNMP
> > > > developers / contributors (primarily Bart Van Assche) to discuss how we
> > > > could best help the project. The meeting went well, at least form our
> > > > perspective.
> > > >
> > > > The pain points we identified were:
> > > >
> > > > * bug mountain
> > > >
> > > > * help users on the mailing list
> > > >
> > > > * patch / MR handling process
> > > >
> > > > * move out of SourceForge
> > 
> > If a release of 5.7.4 is done in the next month or so, that might also be a 
> > good time to put up the "we're moving" sign. If we can get a few people 
> > whose done this before to get together, coming up with the plan could be 
> > done in parallel with the 5.7.4 work. 
> > 
> > > >
> > > > * clean up headers in /include/net-snmp/system/ which are a mess and 
> > > > have
> > > > import loops
> > > >
> > > > * #ifdef hell / too many supported configurations
> > > >
> > > > Keith is currently focused on streamlining the process of handling merge
> > > > requests, which will make it easier to handle the bug mountain. I will
> > > > probably be focusing on the headers for now in hopes that it will make
> > > > other processes easier as well. We can help on the repo move whenever 
> > > > y'all
> > > > are ready to pull the trigger on that.
> > 
> > I have a few questions for the core team regarding this; expect another 
> > email to the mailing list in a little bit.
> > 

=== snip 

-- 
Thanks,
Keith (pantherse)

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Net-snmp-coders mailing list
Net-snmp-coders@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders