EDK 4.4.0.0 with ESI support

[Salvoxia]

The attacker that laid the final blow does neither have a character ID, nor corporation ID, nor alliance ID (or a ship or weapon type for that matter). Nor did they do any damage...
Right now, EDK cannot handle such involved parties..

I'm not sure yet how to handle such kills, but with ESI digging out the old ones, it seems necessary that it can.

[neffets]

as it didnt interrupt the process (newer killmails still beeing fetched) and it was only this one kill i tend to say "ignore these"

[alghara]

I still have some problem with some kill

Error processing involved party. Kill-ID: 23806, KillID = 23806

some idea

[Salvoxia]

Hi,

I'm afraid that's similar to what neffets reported: The one and only attacker on this kill doesn't have a character ID, corporation ID, alliance ID, ship type ID or weapon ID and did no damage.
In its current state, EDK cannot handle something like that , sorry.

Best Regards,
Salvoxia

[alghara]

oki thx.

I have also some issue with the cron. Apparently don't work. Somebody have the same issue ?

Code: Select all

./cron_zkb.php
PHP Deprecated:  Methods with the same name as their class will not be constructors in a future version of PHP; RankModStandardClass has a deprecated constructor in /var/www/html/killboard/mods/rank_mod/class.rankmod.php on line 1691
2018/05/14 12:15:21 - Starting zKB Import

 PHP Notice:  No suitable handler for query found. "select" in /var/www/html/killboard/common/includes/class.dbcachedquery.php on line 263

[Salvoxia]

Hi,

despite the deprecation warning and the notice, can you confirm the cron job does not work?
Looks to me like wrong error reporting settings. These messages would not stop the cron job.

Regarding the deprecation warning: That's caused by the Rank Mod. There's an updated version available for better compatibility with PHP7.x, see this post.

Best Regards,
Salvoxia

[mccrashman]

Hi,

after registering several other SSO keys to my killboards, i have some trouble fetching killmails via ESI. The strange thing is, that the first key works without problems (the one with the registered application).

I've got the following SQL errors:

Code: Select all

SQL execution error: Column 'kll_x' cannot be null
SQL: UPDATE kb3_kills JOIN kb3_mails ON kb3_mails.kll_id = kb3_kills.kll_id SET kb3_kills.kll_external_id = ?, kb3_mails.kll_external_id = ?, kll_modified_time = UTC_TIMESTAMP(), kb3_kills.kll_x = ?, kb3_kills.kll_y = ?, kb3_kills.kll_z = ? WHERE kb3_kills.kll_id = ? AND (kb3_kills.kll_external_id IS NULL OR kb3_kills.kll_x = 0)

I took a look in my webserver error logs. Following error message may be the error location.

Code: Select all

[Tue May 15 11:53:22.664053 2018] [:error] [pid ***] [client ***:53377] PHP 256:  SQL execution error. in /var/www/***/common/includes/class.dbpreparedquery.php on line 96

Any ideas where i can find more information, what went wrong?

Best Regards,
mccrashman

[Salvoxia]

Hi,

this is a known issue and there's a hotfix available.
Please see in the opening post at the bottom in the list of the known issues (called "An SQL error might occur when fetching ancient killmails").

That error occurs when kills are fetched that do not have any coordinates in space, because they are too old. The parser did not expect that the coordinates could be missing.

Best Regards,
Salvoxia

[neffets]

Came upon another error.

Since ESI is now a thing, I thought Zkillboard might have fetched some older kills where my corporation was a party. All works like it should, until ZKB-Cron-Fetch is stuck at some point. The last two times I simply edited the latest kill time, and it worked again for some kills. But now it's stuck again, and I it's hard to always chance last kill timestamp.

cron_zkb.php

Code: Select all

#!/usr/bin/env php 2018/05/15 16:30:25 - Starting zKB Import
2018/05/15 16:30:41 - ZKBApi: https://zkillboard.com/api/corporationID/1639231548/
2018/05/15 16:30:41 - 0 kills were posted and 110 were skipped (200 kills fetched)
2018/05/15 16:30:41 - Timestamp of last kill: 2011-07-04 19:49:00
2018/05/15 16:30:41 - Time taken = 15.182667016983 seconds.

It's simply stuck at this point, each refresh doesn't bring it to a newer kill.


Any Idea?

[Salvoxia]

Well, the last time I heard of such problems was in 2014 Back than the problem was that all kills returned for a fetch cycle had the same timestamp, so fetching would not notice it was actually advancing.

I tried reproducing that behavior, but my test killboard just kept fetching.
Could you tell me what settings you are using for the negative time offset (in ACP -> zKB Fetch)and the maxmimum number of kills to be fetched per cycle (in ACP -> Configuration)? And what is the current "last kill timestamp" setting in ACP -> zKBFetch?

Best Regards,
Salvoxia