2.4 Addrecord() speed

[Russ Tyndall]

On 8/1/06 4:42 PM, "Ruslan Zasukhin" <sunshine at public.kherson.ua> wrote:

> On 8/1/06 10:59 PM, "Russ Tyndall" <fitzbew at nc.rr.com> wrote:
> 
>>> Well, please do next:
>>> 
>>> 
>>> * stop 3 servers
>>> 
>>> * in the ini file set VerboseLevel = 3
>>> 
>>> * run them each after each
>>> 
>>> * run your test with loop.
>>> 
>>> * stop a server
>>> 
>>> Send me please 3 log files of each server. Note that log file show time up
>>> to milliseconds for each command.
>>> 
>>> So we must be able easy compare logs
>>> 
>> 
>> 
>> Ruslan, 
>> 
>> I apologize for not being clear, but I only have 1 copy of vServer running
>> on a X.4.7 machine.  I'm seeing the speed slow-down/diffs while connecting
>> TO that machine FROM other machines on my local network.
>> 
>> This is why I think it is so odd that the machine that hosts vServer is the
>> *slowest*, even though I am running my RB client app ON that machine.
>> 
>> Still, I will follow your instructions but just for 1 vServer.
> 
> Aha, 
> 
> Then only speed of client app itself can affect this.
> Then you need check if DebugLevel is the same ..
> Puzzle. 

I've "played" with this issue for several hours, but still am not sure
what's going on.  I've done testing with 2.4 vs. 2.3 and 2.4 is always
slower just for this rst.Addrecord() procedure which adds about 260 recs.

On 2.3, this activity always takes about 1-2 secs --- on 2.4, the time it
takes can range from 7 seconds to 150 seconds!  This is the time I'm getting
running my RB client on the SAME machine that is running vServer.

All the debuglevel params are off, all the vServer verbosity is set to
lowest setting.

Ruslan, I'll try and get something together for you to look at, but it might
not be until tomorrow.

-- 
Russ Tyndall
Wake Forest, NC