[Forgive the google posting, but my nntp host is dead.]
I'm building (well, built) a new box that is a clean install of FreeBSD
12.2, with a current port of innd.
However starting inn leads to (in a few seconds) a syslog of:
kernel: pid 39316 (innd), jid 0, uid 8: exited on signal 11 (core dumped)
(core dumps are nowhere to be found.)
For comparison, I updated my (previously good innd) 11.4 box (which the
new is to replace) to all currents ports, and I get the same. This even >though the innd version has not changed.
Is there a current problem that I should have heard of?
In article <3a7f139e-98a7-4ac0...@googlegroups.com>,
Drew Lawson <drewla...@gmail.com> wrote:
[Forgive the google posting, but my nntp host is dead.]
I'm building (well, built) a new box that is a clean install of FreeBSD >12.2, with a current port of innd.
However starting inn leads to (in a few seconds) a syslog of:
kernel: pid 39316 (innd), jid 0, uid 8: exited on signal 11 (core dumped)
Are you using clang 11 ?
On Monday, May 10, 2021 at 9:09:50 AM UTC-4, YouKnowWho!! wrote:
In article <3a7f139e-98a7-4ac0...@googlegroups.com>,
Drew Lawson <drewla...@gmail.com> wrote:
[Forgive the google posting, but my nntp host is dead.]
I'm building (well, built) a new box that is a clean install of FreeBSD
12.2, with a current port of innd.
However starting inn leads to (in a few seconds) a syslog of:
kernel: pid 39316 (innd), jid 0, uid 8: exited on signal 11 (core dumped)
Well, a completely clean install (without my config changes and spool) is running fine.
I'm putting my stuff in one piece at a time and hoping it goes well. >'makehistory -O' was also crashing, so I'm guessing something was wonky in my overview.
In article <67680b5d-4578-4506-ba82-bc306630fe84n@googlegroups.com>
Drew Lawson <drewlawson42@gmail.com> writes:
On Monday, May 10, 2021 at 9:09:50 AM UTC-4, YouKnowWho!! wrote:is running fine.
In article <3a7f139e-98a7-4ac0...@googlegroups.com>,Well, a completely clean install (without my config changes and spool)
Drew Lawson <drewla...@gmail.com> wrote:
[Forgive the google posting, but my nntp host is dead.]
I'm building (well, built) a new box that is a clean install of FreeBSD >>> >12.2, with a current port of innd.
However starting inn leads to (in a few seconds) a syslog of:
kernel: pid 39316 (innd), jid 0, uid 8: exited on signal 11 (core dumped) >>
I'm putting my stuff in one piece at a time and hoping it goes well. >>'makehistory -O' was also crashing, so I'm guessing something was wonkyin my overview.
I put back everything in ~news/etc, ~news/db/active, and >~news/spool/articles. Ran 'makehistory -O' and it starts up and
runs just fine.
I knew the overview from a different machine/disk would be invalid,
but I didn't expect it to break things. Something learned for the
month.
Another thing learned is that the core dump was in ~news/spool.
--
Drew Lawson
I only came in search of answers, never planned to sell my soul
I only came in search of something left that I could call my own
In article <67680b5d-4578-4506-ba82-bc306630fe84n@googlegroups.com>
Drew Lawson <drewlawson42@gmail.com> writes:
On Monday, May 10, 2021 at 9:09:50 AM UTC-4, YouKnowWho!! wrote:
In article <3a7f139e-98a7-4ac0...@googlegroups.com>,Well, a completely clean install (without my config changes and spool) is running fine.
Drew Lawson <drewla...@gmail.com> wrote:
[Forgive the google posting, but my nntp host is dead.]
I'm building (well, built) a new box that is a clean install of FreeBSD >>>> 12.2, with a current port of innd.
However starting inn leads to (in a few seconds) a syslog of:
kernel: pid 39316 (innd), jid 0, uid 8: exited on signal 11 (core dumped) >>
I'm putting my stuff in one piece at a time and hoping it goes well.
'makehistory -O' was also crashing, so I'm guessing something was wonky in my overview.
I put back everything in ~news/etc, ~news/db/active, and ~news/spool/articles. Ran 'makehistory -O' and it starts up and
runs just fine.
I knew the overview from a different machine/disk would be invalid,
but I didn't expect it to break things. Something learned for the
month.
Another thing learned is that the core dump was in ~news/spool.
On 5/10/21 8:04 AM, Drew Lawson wrote:
In article <67680b5d-4578-4506-ba82-bc306630fe84n@googlegroups.com>
Drew Lawson <drewlawson42@gmail.com> writes:
Well, a completely clean install (without my config changes and spool) is running fine.
I'm putting my stuff in one piece at a time and hoping it goes well.
'makehistory -O' was also crashing, so I'm guessing something was wonky in my overview.
I put back everything in ~news/etc, ~news/db/active, and
~news/spool/articles. Ran 'makehistory -O' and it starts up and
runs just fine.
I knew the overview from a different machine/disk would be invalid,
but I didn't expect it to break things. Something learned for the
month.
Another thing learned is that the core dump was in ~news/spool.
This sounds like ino64.
When upgrading from FreeBSD 11 to 12, a change in the size of the
ino_t type causes a binary incompatibility in the overview
database
if the default tradindexed format is used.
In article <satmdi$11j1$1@842ffb22-07e1-11e5-a459-00266cf00584.csiph.com>
Kevin Bowling <kevin.bowling@kev009.com> writes:
On 5/10/21 8:04 AM, Drew Lawson wrote:
In article <67680b5d-4578-4506-ba82-bc306630fe84n@googlegroups.com>
Drew Lawson <drewlawson42@gmail.com> writes:
Well, a completely clean install (without my config changes and spool) is running fine.
I'm putting my stuff in one piece at a time and hoping it goes well.
'makehistory -O' was also crashing, so I'm guessing something was wonky in my overview.
I put back everything in ~news/etc, ~news/db/active, and
~news/spool/articles. Ran 'makehistory -O' and it starts up and
runs just fine.
I knew the overview from a different machine/disk would be invalid,
but I didn't expect it to break things. Something learned for the
month.
Another thing learned is that the core dump was in ~news/spool.
This sounds like ino64.
[snip]
When upgrading from FreeBSD 11 to 12, a change in the size of the >> ino_t type causes a binary incompatibility in the overview
database
if the default tradindexed format is used.
That sounds like exactly the problem.
And a type size change makes perfect sense.
I normally only scane UPDATING for base system issues. It didn't
occur to me to check for port issues. I guess I've gotten lucky
with sloppy methods.
Sysop: | Keyop |
---|---|
Location: | Huddersfield, West Yorkshire, UK |
Users: | 379 |
Nodes: | 16 (2 / 14) |
Uptime: | 43:51:36 |
Calls: | 8,141 |
Calls today: | 4 |
Files: | 13,085 |
Messages: | 5,857,952 |