spamassassin-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From James Hsieh <jhs...@stickman-computing.org>
Subject Re: SpamAssassin 3.4.2. - err.h requirement
Date Tue, 18 Sep 2018 02:57:50 GMT
So this is odd:

I finally started this without the daemonize option to see if it gave me anything more enlightening
and got:

Sep 17 22:44:15.326 [21704] error: config: no rules were found! Do you need to run 'sa-update'?
config: no rules were found!  Do you need to run 'sa-update'?

Sure enough /var/lib/spamassassin/3.004002/updates_spamassassin_org is empty.

/usr/local/bin/curl -s -L -O --remote-time -g --max-redirs 2 --connect-timeout 30 --max-time
300 --fail -o MIRRORED.BY -- http://spamassassin.apache.org/updates/MIRRORED.BY <http://spamassassin.apache.org/updates/MIRRORED.BY>

This is what appears to have failed.  It didn't download anything.

superchicken:root# host spamassassin.apache.org
spamassassin.apache.org has address 95.216.24.32
spamassassin.apache.org has address 40.79.78.1
spamassassin.apache.org has IPv6 address 2a01:4f9:2a:185f::2
spamassassin.apache.org mail is handled by 10 mx1-lw-eu.apache.org.
spamassassin.apache.org mail is handled by 10 mx1-lw-us.apache.org.
superchicken:root#

The issue is with the mirror at 95.216.24.32 - it's returning a 404 for the file.  If you
go to 

http://95.216.24.32/updates/MIRRORED.BY <http://95.216.24.32/updates/MIRRORED.BY>

you get a not found.

I copied the MIRRORED.BY file from the 3.004001 directory then re-ran sa-update.  It pulled
down rules and I'm in business.

Kevin, thanks for the patch.  It works.  Sorry for the extra noise and perhaps we need to
do something about that broken mirror?

--James

> On Sep 17, 2018, at 10:17 PM, Bill Cole <bill@scconsult.com> wrote:
> 
> On 17 Sep 2018, at 22:03 (-0400), Kevin A. McGrail wrote:
> 
>> Interesting I think Bill added that code.
> 
> Nope. From svn:
> 
> 3042	mmartinec	810883	    die sprintf("child process [%s] exited or timed out ".
> 3043	 	 	                "without signaling production of a PID file: %s",
> 3044	 	 	                $pid, exit_status_str($child_stat,0)) unless $serverstarted;
> 
>> 
>> On 9/17/2018 9:16 PM, James Hsieh wrote:
>>> This does help things to build.  I can build successfully.  I now have a problem
where spamd seems to be jumping out of the startup wait loop.
>>> 
>>> I keep hitting this code:
>>> 
>>>    die sprintf("child process [%s] exited or timed out ".
>>>                "without signaling production of a PID file: %s",
>>>                $pid, exit_status_str($child_stat,0)) unless $serverstarted;
>>> 
>>> However, there's no waiting for things to start.  I'm investigating this now.
>>> 
>>> --James
>>> 
>>>> On Sep 17, 2018, at 4:53 PM, Kevin A. McGrail <kmcgrail@apache.org>
wrote:
>>>> 
>>>> This patch was added for windows.  Does it help you?  It switches to errno.h
>>>> 
>>>> --- getopt.c    2018-09-13 21:27:52.000000000 -0400
>>>> +++ ../../3.4/spamc/getopt.c    2018-09-17 07:12:38.758722882 -0400
>>>> @@ -21,7 +21,7 @@
>>>> #include <string.h>
>>>> #include <assert.h>
>>>> #include <stdlib.h>
>>>> -#include <err.h>
>>>> +#include <errno.h>
>>>> #include "getopt.h"
>>>> 
>>>> #ifdef WIN32
>>>> @@ -246,7 +246,8 @@
>>>>       if((bp = strchr(longopt, '='))) {
>>>>          opt = strdup(bp+1);
>>>>          if (opt == NULL) {
>>>> -            err(1, NULL);
>>>> +            fprintf(stderr, "%s: %s\n", argv[0], strerror(errno));
>>>> +            exit(1);
>>>>          }
>>>>          longoptlen -= strlen(bp);
>>>>       }
>>>> 
>>>> On 9/17/2018 4:37 PM, James Hsieh wrote:
>>>>> So this isn't a complaint.  I think the SA team has done a great job
getting this release out and yes, it works fine for me under Linux.
>>>>> 
>>>>> However, any creative ways for getting around the fact that err.h is
now part of spamc/getopt.c?  This breaks older OSes (which admittedly I probably shouldn't
care about or use anymore) such as Solaris 10.  For a double dot release, this seems a bit
unfortunate though I agree, really need to get this system on to something more modern.
>>>>> 
>>>>> make -f spamc/Makefile spamc/spamc
>>>>> /usr/local/bin/gcc  -g spamc/spamc.c spamc/getopt.c spamc/libspamc.c
spamc/utils.c \
>>>>>       -o spamc/spamc -L/usr/lib -L/usr/ccs/lib -L/opt/solarisstudio12.3/prod/lib/sparc
-L/opt/solarisstudio12.3/prod/lib -L/lib -L/usr/local/lib  -ldl -lz -lnsl -lsocket
>>>>> spamc/getopt.c:24:17: fatal error: err.h: No such file or directory
>>>>> compilation terminated.
>>>>> *** Error code 1
>>>>> make: Fatal error: Command failed for target `spamc/spamc'
>>>>> Current working directory /export/src/Mail-SpamAssassin-3.4.2
>>>>> *** Error code 1
>>>>> make: Fatal error: Command failed for target `spamc/spamc'
>>>>> 
>>>>> --James
>>>>> 
>>>> -- 
>>>> Kevin A. McGrail
>>>> VP Fundraising, Apache Software Foundation
>>>> Chair Emeritus Apache SpamAssassin Project
>>>> https://www.linkedin.com/in/kmcgrail - 703.798.0171
>>>> 
>> 
>> -- 
>> Kevin A. McGrail
>> VP Fundraising, Apache Software Foundation
>> Chair Emeritus Apache SpamAssassin Project
>> https://www.linkedin.com/in/kmcgrail - 703.798.0171
> 
> 
> 
> -- 
> Bill Cole
> bill@scconsult.com <mailto:bill@scconsult.com> or billcole@apache.org <mailto:billcole@apache.org>
> (AKA @grumpybozo and many *@billmail.scconsult.com addresses)
> Currently Seeking Steadier Work: https://linkedin.com/in/billcole <https://linkedin.com/in/billcole>

Mime
View raw message