Error

[tuffgurlz]

I edited the main text and it was working fine then it took forever to load and gave me this message 1194 - Table 'whos_online' is marked as crashed and should be repaired

 

delete from whos_online where time_last_click < '1217535929'

 

 

[TEP STOP]

 

 

 

then when I refreshed this was the message

 

Fatal error: Call to a member function add_current_page() on a non-object in /home/lindsay/public_html/shop/includes/application_top.php on line 312

[♥geoffreywalton]

Google this

 

Click on Repair table and duck.

[koniosek]

aby idea from where this 'whos_online' error appear ?

We have moved some clients to new server with mysql5 and php5 and from this time we get this error 3-4 times every day on few osc shops.

About month ago we have moved other server to new with same config as above and there is evertything ok.

[Guest]

I got this just the other day. :)

 

It happens when the server crashes or is rebooted. (not specifically sure if it is caused by a forceful or graceful reboot, i honestly do not know the difference)

 

easy fix is logging into phpmyadmin, check mark the table that's acting up & click repair (pull down toward the bottom of the page)

[koniosek]

server, mysql and http have uptime more then 2 days then I don't think this is the problem

[Guest]

not sure what you mean, is it still showing that error? if so, you did not run the repair. it cannot fix itself :)

 

it is normal for the whos_online to crash with a reboot. it's been posted about countless of times.

i think some explanations posted before were along the lines of:

the table was being written to at the time of the connectivity loss which resulted in a partially written table = corrupted

(this would explain why tables you don't normally write to every other second, like "products" also isn't broken)

when i got this the other day, supertracker also suffered the same fate, but it was fixed and working after a repair.

 

 

if the site worked immediately after the reboot, it may have something to do with the sql cache. i do not know much about it, but if you have even a slight idea of how a "cache" works, it would make perfect sense for the error to not show itself immediately. only making an assumption on this part. i've never had that error linger so i've never had to investigate beyond the table repair.

[koniosek]

read carefully ... "error shows 3-4 times every day" then it must be fixed before to appear again ... and yes it still shows few times without restart mysql/server/http

as you say maybe it;s some cache or other memory problem ... we are investigating it for now

but if you have any idea what it could be I will be grateful for any help or clue ...

 

and as you say "it is normal for the whos_online to crash" ... from my experience sessions too

but I'm wonder why osc doesn;t have any autorepair function for this 2 tables ... it will be best option in this case

(or maybe I miss something)

[Guest]

but I'm wonder why osc doesn;t have any autorepair function for this 2 tables

because them crashing regularly (or without a reboot) isn't normal.

 

 

if it's happening without the server rebooting, you're probably losing connectivity to mysql or other integral services that aid in writing to a database somehow. you'd probably want to ask your host if you're hitting any limits that would cause loss of connectivity to anything, for any period of time.

 

assuming you are on shared, your limits can be easily maxed out by other users if the host oversells, or if there is an abusive user/resource hog sharing the server with you.

if it's a shared host, they probably won't tell you if this is happening because of another user, you'll just magically find things working smoother after you complain. you must stress to them that this is not a bug in the code you're using. a lot of hosts try to pass the buck and blame scripts for stuff like this, but that simply isn't the case here, or we'd all be having this problem on a daily basis.