Jason S's questions

I have an LDAP server running using Apache Directory Server 1.5.7. How do I backup & restore the data?

(Apologies if I've got the terminology wrong, I'm fairly new to LDAP)

I am setting up a local LDAP server (Apache Directory Server) with the following structure:

o={my organization name} [objectClass=organization]
  ou=groups [objectClass=organizationalUnit]
    cn=someGroup [objectClass=groupOfUniqueNames]
    cn=otherGroup [objectClass=groupOfUniqueNames]
    ...
  ou=users [objectClass=organizationalUnit]
    cn=user1 [objectClass=inetOrgPerson]
    cn=user2 [objectClass=inetOrgPerson]
    cn=user3 [objectClass=inetOrgPerson]
    ...

I also set up some basic authorization according to the manual.

Everything works great.

Now I have an issue. I have another server running Atlassian Crowd that needs access to this LDAP, and I would like to give that service its own LDAP authorization entry, to partition access rights. But it's not a user, it's a service.

What objectClass is used for service identities?

(and as a newbie to LDAP, how do you find out that groupOfUniqueNames is used for groups, inetOrgPerson is used for user entries? That seems to be the norm.)

I have a site with this existing rule:

RewriteRule ^(.*)$ index.php?q=$1 [L,QSA]

so that it captures all URLs and translates them to this query.

How can I precede this with something that will make an exception for URLs of the form

http://www.example.com/services/{anything}

e.g. any URL that refers to something in a /services directory?

I was noticing my disk space disappearing faster than expected, and finally narrowed it down to a rewrite.log file that was 4 GB in size!

Is there a way to rotate the various Apache logs (rewrite, error, access, etc.) on a Win32 PC so that only the most recent entries are there and I can limit the data size that results? I found the bit about log rotation on Apache's website but it's Unix-centric.

Edit: I got rotatelogs.exe to work, and it's great except that it slows the server response down noticably so I rejected the idea of using it.

I have an application which needs a database + would greatly benefit from a MVCC model. It would ideally support permanent history storage, access of the data at a particular point in time, and incremental backups.

I'm looking at the databases that might support this. I know CouchDB can (aside from the incremental backups, I'm not sure about that) but in this application I really need a relational database.

I'm surprised that MySQL lists MVCC support in Innodb / Falcon. I get the impression that maybe this is only for the database state within a transaction, and once all the transaction levels are committed, the history is lost.

What about PostgreSQL?

I have an .mdb file that a coworker wants me use to convert to tables on our MySQL server. I'm familiar with MySQL, but have very little experience with Access (nor do I really want to learn).

Any suggestions on where to find out the information I need?

I have a Windows XP PC that I'm using as a server for a couple of apps. Is there a program I can use to monitor the memory usage / CPU load? The PC has Apache and PHP available.

(I don't need fast realtime updates, once every 5 seconds would be fine, I just want to see memory usage + CPU load and make sure my server's not overtaxed)

I've never seen phpinfo() give errors before. What gives? What should I look for in php.ini?

Warning: phpinfo() [function.phpinfo]: It is not safe to rely on the system's 
timezone settings. You are *required* to use the date.timezone setting or the 
date_default_timezone_set() function. In case you used any of those methods 
and you are still getting this warning, you most likely misspelled the
timezone identifier. We selected 'America/New_York' for 'EDT/-4.0/DST'
instead in /Users/{me}/Sites/localmirror/htdocs/tests/phpinfo.php on line 3

offending file:

<?php

phpinfo();

?>

edit: Ah, glimmer of understanding, the warning means I need to set "date.timezone" in php.ini. I thought they were talking about "you" the php programmer, instead they're talking about "you" the sysadmin.

Are there any particular reasons why I would not want to use the --deltas argument, either for an incremental or a complete svnadmin dump?

...so I want to back up my Subversion repository and I know how to do that using svnadmin dump.

My question is, what is the best way to verify integrity of the dump files?

I figure I'll use them to restore to a mirror server... but then I'm not sure how to compare the original server with the mirror server.

Dumb question: is there a batch-mode utility (e.g. particular options for curl or wget) that can let me issue raw HTTP commands, and either output the raw results or pipe/redirect it as I see fit?

I know telnet theoretically works but I'm running Windows XP and both Microsoft telnet and PuTTY seem to be interactive and seem to screw things up.

edit: (clarification) I've used curl before, so am familiar w/ the idea, as well as how to send parameters via GET/POST, just not how to control the headers explicitly, and how to keep headers in the output... I can see two possible methods that would be useful to me:

1. Raw HTTP I/O -- I had meant literally like telnet but in batch mode (not opening up a new window or mucking with the terminal settings):

   GET http://someserver/someresource/blah.html
   HTTP/1.1 301 Moved Permanently
   Date: Wed, 14 Oct 2009 21:23:40 GMT
     ...
   

2. "easy" input, raw HTTP output -- using curl in the normal methods, but being able to see the raw output verbatim:

   > curl --some-option --another-option http://someserver/someresource/blah.html
   HTTP/1.1 301 Moved Permanently
   Date: Wed, 14 Oct 2009 21:23:40 GMT
     ...
   

I have Tomcat 6.0 up and running. I went to tweak the memory sizes and realized that I have it running on the Sun JDK 1.6 client JVM. I don't have the Sun server JVM installed:

C:\>java -client -version
java version "1.6.0_16"
Java(TM) SE Runtime Environment (build 1.6.0_16-b01)
Java HotSpot(TM) Client VM (build 14.2-b01, mixed mode, sharing)

C:\>java -server -version
Error: no `server' JVM at `C:\appl\java\jre6u16\bin\server\jvm.dll'.

To clarify: I know how to switch JVMs in Tomcat from client to server. I just have to pick the most appropriate server JVM.

Am I going to notice a big difference between client and server JVMs? If I want the Sun server JVM, do I have to reinstall the whole JDK? Are there other free server alternatives that would be good for running Tomcat?

So I just got JIRA running with Crowd. How do I allow users to change their own passwords? I must be missing something obvious.

I have a test account, I can login to JIRA but I don't see any way to change my password, and I can't login to Crowd via that test account.

I'm evaluating JIRA for task/bug tracking. Looks good so far but it keeps reporting URLs with "localhost" rather than the server name in outgoing email messages. (example below) Is there a way to fix this?

dollar bills wear out too fast

             Key: TWEEDLEDEE-2
             URL: http://localhost:8082/browse/TWEEDLEDEE-2
         Project: Tweedledee
      Issue Type: Whiz-bang-thingamajig!
      Components: Dee
     Environment: USA money
        Reporter: Jason S (admin)
        Assignee: Jason S

dollar bills just wear out too fast. find a solution please.

-- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://localhost:8082/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

How do I figure out which ./configure options to use when building PHP 5.3.0 on Mac OS X? There are 10 zillion options and I have no idea which ones to use.

All I want is to make sure that I have PDO and PDO+MySQL enabled.

[[[[scream!]]]] I ran sudo port install php5-mysql and it "worked", but when I run php:

$ php
PHP Warning:  Cannot load module 'mysql' because required module 'mysqlnd' is not loaded in Unknown on line 0

Warning: Cannot load module 'mysql' because required module 'mysqlnd' is not loaded in Unknown on line 0
dyld: lazy symbol binding failed: Symbol not found: __mysqlnd_palloc_rinit
  Referenced from: /opt/local/lib/php/extensions/no-debug-non-zts-20090626/mysqli.so
  Expected in: dynamic lookup

dyld: Symbol not found: __mysqlnd_palloc_rinit
  Referenced from: /opt/local/lib/php/extensions/no-debug-non-zts-20090626/mysqli.so
  Expected in: dynamic lookup

Trace/BPT trap

aha! The maintainer of macports found the same problem. I did sudo port selfupdate and reran the install of php5 and it worked fine.

I'm trying to configure PHP on Mac OS X (10.5.7) and am really confused as to where this program "lives".

Here's what I've found / not found:

php: /usr/bin/php
php.ini: /etc/php.ini
extensions directory: /usr/lib/php/extensions

Are there other places where PHP components are located?

I'm running OS X 10.5; it looks like it came with Apache and PHP installed (minus some minor configurations which I turned on per this page; I've used Apache before so I know the basics of how httpd.conf works).

I've got a pre-existing script which uses PDO. I've got a MySQL database and can easily configure my script to access the database via PDO MySQL or PDO ODBC. The problem is, that even though I enabled the PDO MySQL and PDO ODBC extensions in php.ini, phpinfo() reports the only PDO drivers are sqlite2 and sqlite. I'm guessing the relevant extension .dll or .so files are not present? How do I get them?

note: I'm using the built-in install for PHP. (see apple's page on enabling php, which doesn't say anything about configure or adding additional .so files)

Anyone know how you get a client to switch license servers? I just installed R2009a and typed in the wrong license server (we still have an old server running R2006, I pointed to that by mistake); I really don't want to uninstall/reinstall the whole @#$@#$ program again.