If you have changed your indices and you had some data already in
database,you should then re-index your database using 'slapindex' command
after stopping slapd, and then restart slapd.

--Qutub

> Wouldn't it also be effective to index the uid field?  Especially in
> this case where the users are popping there account's every minute.
>
> Also a quick question I added the indexes you provided to my slapd.conf
>  and restarted slapd. After this my qmail-ldap couldn't lookup any user
>  via mail/mailAlternateAddress but could by uid.  Do the indexes need
> to  be setup before any data is entered into the LDAP  or should the
> indexes  be created from existing data?
>
> --Mike
>
> Oscar Bonilla wrote:
>
>> Ohh sorry! I should have seen this before. The problem is probably
>> that you didn't create any indexes on the LDAP database. Therefore
>> your slapd processes are getting stuck trying to fetch the info.
>>
>> Modify your slapd.conf and add the lines
>>
>> index   objectClass                                     eq
>> index   mail,mailAlternateAddress                       eq
>> index   cn,givenName,sn                                 eq,sub
>> index   default                                         sub
>>
>> to the end of the file
>>
>> regards,
>>
>> -Oscar
>>
>> On Tuesday, August 27, 2002, at 09:36  AM, Oscar Bonilla wrote:
>>
>>> Try running a vmstat with a delay of 1 or 2 seconds and send the
>>> output.
>>>
>>> Which OS is this? What's the speed of the processors?
>>>
>>> How are you doint the delivery? using procmail? with a program you
>>> wrote?
>>>
>>> are you running an anti-virus?
>>>
>>> regards,
>>>
>>> -Oscar
>>>
>>> On Tuesday, August 27, 2002, at 09:08  AM, Matt wrote:
>>>
>>>> What does a �normal� mail server with about 15,000 users look like
>>>> load wise?
>>>>
>>>>
>>>>
>>>> I�ve only got about 30 users on here right now and had them popping
>>>> their messages 1 once a minute and had another mail server sending a
>>>>  ton of messages in and this is what the top display looked like:
>>>>
>>>>
>>>>
>>>> 11:05am  up 4 days, 0 min,  1 user,  load average: 10.22, 8.70, 6.50
>>>>
>>>> 79 processes: 72 sleeping, 7 running, 0 zombie, 0 stopped
>>>>
>>>> CPU states: 93.6% user,  6.3% system,  0.0% nice,  0.0% idle
>>>>
>>>> Mem:   255656K av,  246580K used,    9076K free,       0K shrd,
>>>> 75552K buff
>>>>
>>>> Swap:  522104K av,       0K used,  522104K free
>>>>       120736K cached
>>>>
>>>>
>>>>
>>>>   PID USER     PRI  NI  SIZE  RSS SHARE STAT %CPU %MEM   TIME
>>>>   COMMAND
>>>>
>>>>  7390 root      39   0  6640 6636  1952 R    11.3  2.5  1:01slapd
>>>>
>>>>  9949 root      39   0  6640 6636  1952 S    11.3  2.5   0:41 slapd
>>>>
>>>>  5664 root      39   0  6640 6636  1952 S     9.3  2.5  2:41slapd
>>>>
>>>>  5661 root      39   0  6640 6636  1952 S     7.9  2.5  3:45slapd
>>>>
>>>> 11493 root      39   0  6640 6636  1952 S     7.9  2.5   0:33 slapd
>>>>
>>>> 12626 root      39   0  6640 6636  1952 R     7.9  2.5   0:24 slapd
>>>>
>>>>  7230 root      39   0  6640 6636  1952 R     7.5  2.5  1:03slapd
>>>>
>>>>  7231 root      39   0  6640 6636  1952 S     7.1  2.5  1:10slapd
>>>>
>>>> 12512 root      16   0  6640 6636  1952 S     6.1  2.5   0:20 slapd
>>>>
>>>>  7238 root      39   0  6640 6636  1952 S     5.5  2.5  1:01slapd
>>>>
>>>>  5825 root      39   0  6640 6636  1952 R     3.7  2.5  1:38slapd
>>>>
>>>>  7458 root      39   0  6640 6636  1952 R     1.9  2.5   0:59 slapd
>>>>
>>>>  7250 root      16   0  6640 6636  1952 S     0.7  2.5  1:10slapd
>>>>
>>>> 18352 matth     15   0  1056 1056   836 R     0.3  0.4   0:00 top
>>>>
>>>> 18548 root      16   0  1324 1324  1164 S     0.3  0.5   0:00
>>>> qmail-lspawn
>>>>
>>>>  6846 root      15   0  1988 1988  1684 R     0.1  0.7   0:00 sshd
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> Is that normal?
>>>>
>>>>
>>>>
>>>> ~ Matt
>>>>
>>>>
>> --
>> pgp fingerprint: BC64 2E7A CAEF 39E1 9544  80CA F7D5 784D FB46 16C1


-- 
Syed Qutubuddin Ahmad
Network Engineer
CubeXS Private Limited.



Reply via email to