http://bugzilla.lysator.liu.se/show_bug.cgi?id=1071
Summary: Remove the nlinks field of isc_scb
Product: lyskomd
Version: 2.0.7
Platform: All
OS/Version: All
Status: NEW
Severity: trivial
Priority: P2
Component: other
AssignedTo: ceder(a)lysator.liu.se
ReportedBy: ceder(a)lysator.liu.se
QAContact: lyskomd-qa(a)lists.lysator.liu.se
The nlinks field of an isc_scb serves no purpose, unless I'm mistaken.
Get rid of it.
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.
http://bugzilla.lysator.liu.se/show_bug.cgi?id=1069
Summary: $any_float accepts too much
Product: lyskomd
Version: 2.0.7
Platform: All
OS/Version: All
Status: NEW
Severity: minor
Priority: P2
Component: testsuite
AssignedTo: ceder(a)lysator.liu.se
ReportedBy: ceder(a)lysator.liu.se
QAContact: lyskomd-qa(a)lists.lysator.liu.se
The $any_float regexp accepts too much. It could be a lot more restrictive.
However, my attempt to write a better regexp doesn't work. See FIXME in
src/server/testsuite/config/unix.exp.
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.
http://bugzilla.lysator.liu.se/show_bug.cgi?id=563
------- Additional Comments From ceder(a)lysator.liu.se 2003-07-27 16:57 -------
The problems in isc.h mentioned in comment 22 is now finally fixed.
This bug should have remained open until now. :-)
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.
http://bugzilla.lysator.liu.se/show_bug.cgi?id=1065
ceder(a)lysator.liu.se changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|ASSIGNED |RESOLVED
Keywords|inpro |
Resolution| |FIXED
------- Additional Comments From ceder(a)lysator.liu.se 2003-07-26 23:00 -------
...but clamping it to an hour works, and has no bad side effects. This
fix is now also included in the liboop CVS repository. See line 321/338 of
http://cvs.ofb.net/gale/liboop/sys.c.diff?r1=1.16&r2=1.17&sortby=date
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.
http://bugzilla.lysator.liu.se/show_bug.cgi?id=1068
Summary: Bad randomness for password salt
Product: lyskomd
Version: 2.0.7
Platform: All
OS/Version: All
Status: NEW
Severity: minor
Priority: P3
Component: server
AssignedTo: ceder(a)lysator.liu.se
ReportedBy: ceder(a)lysator.liu.se
QAContact: lyskomd-qa(a)lists.lysator.liu.se
We feed the random number generator via a simple
srand(time(NULL) + getpid());
and that is cryptographically very poor. Use /dev/random if it exists.
This random number generator is later used to create a salt for crypt().
See FIXME in ramkomd.c.
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.
http://bugzilla.lysator.liu.se/show_bug.cgi?id=1065
------- Additional Comments From ceder(a)lysator.liu.se 2003-07-16 01:41 -------
!Calle Englund says:
> Efter en liten dykning i kernelsåsen framgår att FreeBSD verkar ha en
> begränsning på tv_sec<=100000000
So, clamping the value to 3600 is more restrictive than necessary.
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.
http://bugzilla.lysator.liu.se/show_bug.cgi?id=1066
Summary: Test suite produces core file
Product: lyskomd
Version: 2.0.7
Platform: All
OS/Version: All
Status: NEW
Severity: trivial
Priority: P2
Component: testsuite
AssignedTo: ceder(a)lysator.liu.se
ReportedBy: ceder(a)lysator.liu.se
QAContact: lyskomd-qa(a)lists.lysator.liu.se
The test suite produces a core file, since the server calls restart_kom() when
it attempts to read the file "nosuchfile".
That kind of error messages should use a new function, exit_kom(), that
does not end with an abort() call.
The test suite should remove core files that it expects to produce. It
should periodically check for the appearance of other core files, and
log a failure if it finds one.
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.
http://bugzilla.lysator.liu.se/show_bug.cgi?id=998
ceder(a)lysator.liu.se changed:
What |Removed |Added
----------------------------------------------------------------------------
Severity|normal |enhancement
Priority|P2 |P5
Target Milestone|--- |Future
------- Additional Comments From ceder(a)lysator.liu.se 2003-07-14 10:22 -------
We have a very real issue in that the database is stretched to its limit.
It takes several hours to run "garb -g" to compact the text file. If more
texts existed, it would take an even longer time.
Until the text compaction happens while the server is on-line, I don't
want to change this.
A particular lyskomd installation can change the "Default garb nice:"
setting in etc/config.
If somebody writes a chapter on "Settings you might want to change" I'd
be glad to include it in lyskomd.texi.
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.
http://bugzilla.lysator.liu.se/show_bug.cgi?id=627
------- Additional Comments From ceder(a)lysator.liu.se 2003-07-14 09:17 -------
adns may open a TCP connection if necessary. Can it open more than one?
We need to investigate the file descriptor use of the library. We might
have to limit the number of pending connects. Consider PROTECTED_FDS - we
might have to update it when integrating adns.
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.
http://bugzilla.lysator.liu.se/show_bug.cgi?id=108
ceder(a)lysator.liu.se changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|ASSIGNED |RESOLVED
Resolution| |FIXED
Summary|Run the garb at a fixed time|Run the garb at a varying
|of day |time of day
------- Additional Comments From ceder(a)lysator.liu.se 2003-07-14 00:26 -------
We now delay 24 hours between each garb run. That is probably good
enough. Also, since the garb only runs when idle, it should do little
harm if it is run during peak hours.
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.