Sudowned's questions

I am attempting to implement Yubikey auth via SSH. I've edited my /etc/pam.d/sshd file as follows, and it seems to work while connecting locally (ssh user@localhost):

#%PAM-1.0
auth       required pam_yubico.so id=20682 authfile=/etc/yubikey_mappings debug trace
auth       required pam_sepermit.so
auth       substack     password-auth
auth       include      postlogin
account    required     pam_nologin.so
account    include      password-auth
password   include      password-auth
# pam_selinux.so close should be the first session rule
session    required     pam_selinux.so close
session    required     pam_loginuid.so
# pam_selinux.so open should only be followed by sessions to be executed in the user context
session    required     pam_selinux.so open env_params
session    optional     pam_keyinit.so force revoke
session    include      password-auth
session    include      postlogin

However, when I try to connect from a remote machine, it does not prompt me for my key. What is the likely issue?

On Wednesday we had an unrelated email issue with our host, and they made some changes which included removing our domain, foo.com, from /etc/local_domains.

This had some far-reaching (and difficult to diagnose) implications, and we didn't actually notice that mails weren't being delivered at all until Thursday. I spent Thursday mistakenly trying to figure out what was wrong with our email cronjobs, but on Friday I finally found the real problem and fixed it.

The issue now is that there's a lot of email missing from the last two days. I need to find those missing messages if at all possible, but unfortunately googling for "courier where do failed emails go" results in a bunch of news stories about meatspace shipments. I've checked inside /var/mail but the files within are empty - I guess cPanel or Courier does things I don't expect with email.

My question is: Where would the undeliverable mails from that period go? Is there any possibility they still exist somewhere and I can recover them?

We've got an unusual mail setup - Google Apps/Gmail in front, with Exim running on our webserver. Mail for any account that doesn't exist in Google Apps is forwarded to Exim. Both of these use the same domain.

Problem is, all mail sent from the Exim accounts (or our PHP app) is only delivered locally, not sent through Gmail.

To perhaps illustrate it better, see this high-quality chart:

[Google Apps] (Tom, Dick, Harry)

   |
   v
[Exim] (Jane, Mary, Sue)

Mail sent to Jane gets forwarded to Exim by Google. Mail sent to Tom is simply delivered by Google. This is the part that works - the problem is, Jane can't email Tom because as far as Exim is concerned, it runs the whole show.

I'm reading about setting up a SmartHost in Exim, but that requires authentication - and critically, the accounts in Exim do not match those in Apps.

How can I configure Exim to deliver local mail via external routes?

A webserver (Apache + MySQL) locked so badly that I had to powercycle it, and immediately logged in to see what was going on afterward. Load average was over 3%, with MySQL hovering between 20% and 80% of the CPU. Curious, I ran SHOW PROCESSLIST; during a spike of activity, only to find...

+-------+--------+-----------+--------+---------+------+-------+------------------+
| Id    | User   | Host      | db     | Command | Time | State | Info             |
+-------+--------+-----------+--------+---------+------+-------+------------------+
| 10762 | root   | localhost | NULL   | Query   |    0 | NULL  | SHOW PROCESSLIST |
| 11754 | phorum | localhost | phorum | Sleep   |    0 |       | NULL             |
| 11755 | phorum | localhost | phorum | Sleep   |    0 |       | NULL             |
| 11756 | phorum | localhost | phorum | Sleep   |    0 |       | NULL             |
| 11758 | phorum | localhost | phorum | Sleep   |    0 |       | NULL             |
| 11759 | phorum | localhost | phorum | Sleep   |    0 |       | NULL             |
| 11761 | elba   | localhost | elba   | Sleep   |    0 |       | NULL             |
| 11763 | phorum | localhost | phorum | Sleep   |    0 |       | NULL             |
| 11764 | phorum | localhost | phorum | Sleep   |    0 |       | NULL             |
| 11765 | phorum | localhost | phorum | Sleep   |    0 |       | NULL             |
| 11766 | phorum | localhost | phorum | Sleep   |    0 |       | NULL             |
| 11767 | phorum | localhost | phorum | Sleep   |    0 |       | NULL    

Just for kicks, I enabled the slow query log and there's nothing there, either. Load average (as well as MySQL's resource usage) has dropped back to normal levels since, but it's still unclear what happened.

Any insight as to what happened? Failing that, what should I log to help get to the bottom of it next time?

After setting up a second network adapter in the process of adding some SSL'd sites to ISS, my instance of SQL Server 2005 is no longer accessible - including through SQL Server Management Studio.

The specific error I receive is:

Cannot connect to xxxxxx.
Additional information:

A connection was successfully established with the server, but then an error occurred during the pre-login handshake. (provider: Shared Memory Provider, error: 0 - No process is on the other end of the pipe.) (Mycrosoft SQL Server, Error: 233)

I've verified the following:

• TCP, Shared Memory, and Named Pipe connections are enabled in Configuration Manager
• Connections to anywhere are enabled in Surface Area Configuration
• Surface Area Configuration didn't list my account as having admin credentials; however, when it tried to connect to the server to fix that the connection failed with the same error as Server Management Studio generated. This unfortunately rules out the fix from 203541.

Very little of what I managed to Google pertained to local connections, and what remained wasn't helpful. Is there any way I can fix this, perhaps by running the server with authentication off long enough to fix my permissions a la MySQL?

Any ideas?

I'm having the weirdest problem with a previous developer's URL rewrites after moving this custom CMS to a new server. Observe this snippet of the .htaccess file:

RewriteEngine On

# If the URL doesn't end in a slash, redirect them to the equivalent that does (for canonical/SEO purporses)
RewriteCond %{REQUEST_URI} !.*/$
RewriteCond %{DOCUMENT_ROOT}%{REQUEST_URI} !-f
RewriteRule ^(health|nutrition|fitness|directory|coupons|download|kids|faq|signup|menu|snp|myaccount|myschool|printmenu)(.*)$ /$1$2/ [R=301,L,QSA,NS]

# Health Pages
RewriteRule ^health/$ health.php [L]
RewriteRule ^health/([^/]+)/$ health.php?title=$1 [L]
RewriteRule ^health/([^/]+)/([^/]+)/$ health_article_details.php?cat=$1&title=$2 [L]

The first part of the Health rewrite:

# Health Pages
RewriteRule ^health/$ health.php [L]

works as expected, but:

RewriteRule ^health/([^/]+)/$ health.php?title=$1 [L]
RewriteRule ^health/([^/]+)/([^/]+)/$ health_article_details.php?cat=$1&title=$2 [L]

don't forward the rest of the URL segment as an argument. I threw a print_r($_REQUEST); in at the start of health.php and with a URL of http://www.foo.com/health/a/b/, I get this output:

Array ()

Why is this happening instead of:

Array ( [title] => a )

Any suggestions would be very appreciated!

I have just finished moving a WordPress site to a new server. Absolutely no changes have been made to the code, but when I try to view the site

mod_fcgid: stderr: PHP Parse error:  syntax error, unexpected '}' in /home/crandall/www/website.com/wp-content/plugins/auto-featured-image/auto-featured-image.php on line 353

I don't see any such issue in the code, which I've pastebin'd below for reference; I haven't been able to find any occurrence of this issue in Google. The only thing I can think of is some extremely esoteric issue relating to PHP versions, as the original server was running PHP 5.2 and the new one is running 5.3.

http://pastebin.com/fLuFVEtB

I'm a little baffled by this - I set up Nagios3's server process on one machine (tala) and NPRE on two others (medea and iiyanara.), and thought it was set up correctly. For various frustrating reasons I had to rebuild medea from a disk image, and as a result NPRE isn't installed anymore.

Imagine my surprise when Nagios is (days later) continuing to merrily report not only that my server is up, but also a series of performance metrics like total processes, etc. As it turns out, I'm pretty sure the Nagios instance has actually been reporting on localhost for both medea and iiyanara all along.

Here's the initial block + first service block from medea.conf:

define host{
    use                     generic-host            ; Name of host template to use
    host_name               medea.phyre.im
    alias                   Medea
    address                 97.107.128.112
    }

define service{
    use                             generic-service         ; Name of service template to use
    host_name                       medea.phyre.im
    service_description             Disk Space
    check_command                   check_all_disks!20%!10%
}

Curiously, if I specify a different IP, Nagios reports that the machine is down. So either:

1. Nagios is using black sorcery to commune with medea despite a total absence of the NPRE service
2. Nagios is reporting on localhost for some reason

...And then it STOPS doing those things if I specify, for instance, 197.107.128.112 as the target IP.

My question is basically this: what the hell?

I'm seeing some intermittent CPU usage spikes to 100%, sort of correlated to these log entries:

[27-Feb-2012 13:29:29] PHP Fatal error:  PHP Startup: apc_mmap: mmap failed: in Unknown on line 0
[27-Feb-2012 13:29:30] PHP Fatal error:  PHP Startup: apc_mmap: mmap failed: in Unknown on line 0
[27-Feb-2012 13:29:31] PHP Fatal error:  PHP Startup: apc_mmap: mmap failed: in Unknown on line 0
[27-Feb-2012 13:29:31] PHP Fatal error:  PHP Startup: apc_mmap: mmap failed: in Unknown on line 0

phpinfo() indicates that APC is set up, and as far as I can tell this error doesn't cause visible 500 errors on the live site, which is a WordPress install that gets about 600k views monthly. Google's been unhelpful so far, and I was hoping that someone here had some insight as to what's causing this and how to fix it. Curiously, this error only shows up /usr/local/apache2/logs/error_log and not the error_log for the cpanel-configured site.

I'm not absolutely certain whether this should be here or on Web Applications, but here goes.

I need wiki software, preferably something that runs on PHP but this is not a hard requirement, which supports namespaces. That is, I need to be able to give users an arbitrary number of wiki spaces that do not overlap or interact with each other. FOSS is preferred as well.

I've been unable to find something like this so far, so I may have to roll my own solution. Does anyone know of such an application?