schestowitz | Wait, I'll try something too | Feb 01 12:15 |
schestowitz | Here's the thing | Feb 01 12:15 |
schestowitz | It seems like the original DB had no users | Feb 01 12:15 |
oiaohm2 | Repair tries to rebuild the database struct. | Feb 01 12:15 |
schestowitz | Like the users that was in it just isn't registered with it | Feb 01 12:15 |
oiaohm2 | Ok the user file of the database could be gone. | Feb 01 12:15 |
schestowitz | *user | Feb 01 12:15 |
oiaohm2 | That might be the extent of the failure. | Feb 01 12:16 |
schestowitz | Should I try to reattach the user to the old DB? | Feb 01 12:16 |
oiaohm2 | I normally like backups on the damaged database before playing any more. IE file backups not mysql backups. | Feb 01 12:16 |
oiaohm2 | Rule of recovering data don't edit it without a backup so you can undo your edit. | Feb 01 12:17 |
schestowitz | Yes, I try | Feb 01 12:17 |
oiaohm2 | Basically get it up temp then talk to host. | Feb 01 12:17 |
oiaohm2 | I guess you don't have access to the raw mysql directories to back that up. | Feb 01 12:18 |
schestowitz | Here is the thing | Feb 01 12:18 |
schestowitz | I'm glad you are here to help me | Feb 01 12:18 |
schestowitz | >> boycottn_wrdp1 81.41 MB | Feb 01 12:18 |
schestowitz | That's the old database | Feb 01 12:18 |
oiaohm2 | The older one should be twice the size at least. | Feb 01 12:18 |
schestowitz | It has no user attached to it, which is strange | Feb 01 12:18 |
schestowitz | OK, GOOD | Feb 01 12:19 |
schestowitz | I'm relived! | Feb 01 12:19 |
schestowitz | Cause for a moment I thought data was maybe lost | Feb 01 12:19 |
oiaohm2 | Ie new one has not created its backup yet. | Feb 01 12:19 |
schestowitz | >> boycottn_wrdp2 37.06 MB | Feb 01 12:19 |
oiaohm2 | So it should expand too. | Feb 01 12:19 |
schestowitz | This is the one I just restored | Feb 01 12:19 |
oiaohm2 | When that part of the mysql cycle comes around. | Feb 01 12:19 |
schestowitz | It obviously has no user attached to it yet | Feb 01 12:19 |
schestowitz | My question is this: | Feb 01 12:19 |
*amarsh04 (n=amarsh04@ppp121-45-62-63.lns11.adl2.internode.on.net) has joined #boycottnovell | Feb 01 12:20 |
schestowitz | is it possible that the old DB was never gone and just the user detached from it? | Feb 01 12:20 |
schestowitz | Should I try to reattach it to the old DB? | Feb 01 12:20 |
schestowitz | And see what happens? | Feb 01 12:20 |
oiaohm2 | If its not just that damaged doing that will do more damage. | Feb 01 12:20 |
oiaohm2 | You need the file backups done on there end first. | Feb 01 12:21 |
schestowitz | So should I not reattack user to the old DB? | Feb 01 12:21 |
oiaohm2 | Not at this time. | Feb 01 12:21 |
schestowitz | Som you advise me to just leave it | Feb 01 12:21 |
schestowitz | Should I attach it to the new DB in the mean time? | Feb 01 12:21 |
oiaohm2 | Until you can get the host to back it up filebase. | Feb 01 12:21 |
schestowitz | So that I can serve pages from the old DB? | Feb 01 12:21 |
oiaohm2 | Yep attached the new database to cover the gap. | Feb 01 12:22 |
schestowitz | Thanks! | Feb 01 12:22 |
schestowitz | Wait.. | Feb 01 12:22 |
schestowitz | Attacch old user or create new one? | Feb 01 12:22 |
oiaohm2 | And most likely put up post telling users that there is a issue. | Feb 01 12:22 |
schestowitz | Yes | Feb 01 12:22 |
oiaohm2 | Is old user used by other database as well. | Feb 01 12:23 |
schestowitz | No, just the broken DB | Feb 01 12:23 |
schestowitz | Just one DB | Feb 01 12:23 |
schestowitz | But it's not attached to the old DB for some reason | Feb 01 12:23 |
oiaohm2 | I would create a new one. basically try not to over lap with anything that could explain the problem. | Feb 01 12:23 |
schestowitz | OK | Feb 01 12:23 |
oiaohm2 | Its expermenting without backups that normally make datarecovery down the track impossiable. | Feb 01 12:24 |
oiaohm2 | Creating new user and new database is not expermienting with your data. | Feb 01 12:25 |
oiaohm2 | Next time server is intermiting increase you backup cycle too schestowitz | Feb 01 12:26 |
oiaohm2 | You have backup the wiki? | Feb 01 12:26 |
schestowitz | Yes | Feb 01 12:27 |
schestowitz | oiaohm2: How frequent? | Feb 01 12:27 |
schestowitz | I did one at 7am today | Feb 01 12:27 |
oiaohm2 | How much are you prepaird to loss. | Feb 01 12:28 |
oiaohm2 | That is the factor. Shortest time is dayly. | Feb 01 12:28 |
schestowitz | I only wrote about 7 posts since the last backup | Feb 01 12:28 |
schestowitz | Hmm.. | Feb 01 12:29 |
schestowitz | http://boycottnovell.com/2009/02/01/androi... WordPress database error: [Table 'boycottn_wrdp2.wp_users' doesn't exist] | Feb 01 12:29 |
schestowitz | SELECT * FROM wp_users WHERE user_login = 'schestowitz' | Feb 01 12:29 |
schestowitz | It must be trying the old table names... | Feb 01 12:30 |
schestowitz | This one might have boycottn_wrdp1 <-- 1 | Feb 01 12:30 |
schestowitz | Let me check | Feb 01 12:30 |
oiaohm2 | At least its displaying something now. | Feb 01 12:32 |
oiaohm2 | Ok complainign about wp_users table being missing. | Feb 01 12:32 |
oiaohm2 | In these problems something is better than nothing. | Feb 01 12:33 |
schestowitz | Let me check mysqladmin | Feb 01 12:34 |
schestowitz | I'll check the structure of it | Feb 01 12:35 |
schestowitz | wp_categories wp_comments & nbsp;wp_crossreferences wp_linkcategories wp_links wp_options wp_pollsa wp_pollsip wp_pollsq wp_post2cat wp_postmeta wp_posts | Feb 01 12:36 |
schestowitz | None for users? | Feb 01 12:36 |
oiaohm2 | That is exactly what the php told you. | Feb 01 12:37 |
oiaohm2 | It cannot find the users table at all. | Feb 01 12:37 |
schestowitz | There seem to be too few posts as well. | Feb 01 12:37 |
oiaohm2 | Now I don't know how you backed it up with it. | Feb 01 12:37 |
schestowitz | wp_posts had 3,846 records | Feb 01 12:37 |
schestowitz | Let me check other backups for comparison | Feb 01 12:38 |
oiaohm2 | Seams like your backup is not in perfect condition either. | Feb 01 12:38 |
oiaohm2 | You are really having a bad run. | Feb 01 12:38 |
schestowitz | I have backups as HTML | Feb 01 12:38 |
schestowitz | But those are not needed since the host has BUs | Feb 01 12:39 |
oiaohm2 | Yep the web equal to paper backups. | Feb 01 12:39 |
schestowitz | Yesterday it was 19.6MB, today 19.7MB | Feb 01 12:39 |
schestowitz | So the backup was consistent | Feb 01 12:40 |
schestowitz | Might this be related to DB size? | Feb 01 12:40 |
schestowitz | Here's the thing | Feb 01 12:40 |
schestowitz | Comment count seems correct.. | Feb 01 12:40 |
oiaohm2 | Wonder if your backup was just on the edge of the splat. | Feb 01 12:41 |
oiaohm2 | Or the start of the splat. | Feb 01 12:41 |
oiaohm2 | Do you have a local mysql database server to restore into Yesterday one. | Feb 01 12:42 |
schestowitz | You mean mysqldump? | Feb 01 12:42 |
schestowitz | oiaohm2: the backup done with the cpanel front end is weird | Feb 01 12:42 |
schestowitz | It echoes everything it resores | Feb 01 12:42 |
schestowitz | So maybe it doesn't restore it entirely. | Feb 01 12:42 |
schestowitz | The page got HUGE... which freezes the browser | Feb 01 12:43 |
schestowitz | Wait | Feb 01 12:43 |
schestowitz | I'm looking at other WP blogs of mine | Feb 01 12:43 |
schestowitz | They have 12 tables | Feb 01 12:43 |
schestowitz | Like in this cas | Feb 01 12:43 |
schestowitz | But different ones. | Feb 01 12:44 |
schestowitz | wp_comments wp_links wp_options wp_postmeta wp_posts wp_terms wp_term_relationships wp_term_taxonomy wp_usermeta wp_users wp_wp2pdf wp_wp2pdf_fonts | Feb 01 12:44 |
oiaohm2 | Notice wp_users | Feb 01 12:44 |
oiaohm2 | Other one is missing it. | Feb 01 12:44 |
schestowitz | Yes | Feb 01 12:44 |
schestowitz | But it misses parts of wp-posts I think... count is too low | Feb 01 12:45 |
schestowitz | So maybe I should resore from command line | Feb 01 12:45 |
schestowitz | But the host gives no shell | Feb 01 12:45 |
schestowitz | I might have to use cron job in cpanel | Feb 01 12:45 |
oiaohm2 | Normal | Feb 01 12:45 |
schestowitz | Oh. | Feb 01 12:45 |
oiaohm2 | Issue here give you shell you can do major harm. | Feb 01 12:46 |
oiaohm2 | Reason why you have to wait to do some particular commands. | Feb 01 12:46 |
oiaohm2 | I love when person can restore locallly. | Feb 01 12:46 |
oiaohm2 | You have mysqladmin access if we can see yesterdays and todays compare most likely can add the differences back in. | Feb 01 12:47 |
oiaohm2 | Either todays did not upload right or its never backed up right because the problem was starting. | Feb 01 12:47 |
schestowitz | It increased in size over time though | Feb 01 12:47 |
oiaohm2 | Lets try to work out what one. | Feb 01 12:47 |
schestowitz | So it's not always cut | Feb 01 12:48 |
oiaohm2 | Increasing size does not mean that the data in there is perfect. | Feb 01 12:48 |
schestowitz | DB sizes: boycottn_wrdp1 81.41 MB vs boycottn_wrdp2 37.06 MB | Feb 01 12:48 |
oiaohm2 | We focus on the backups at this point we cannot backup boycottn_wrp1. | Feb 01 12:49 |
oiaohm2 | Do you have a local running mysql ? | Feb 01 12:49 |
schestowitz | I think that Shane made a lovcal full backup a month ago | Feb 01 12:50 |
schestowitz | Me? No... no local sql server on my PC | Feb 01 12:50 |
schestowitz | Let me try to upload and rewrite db2 (the one from backup) | Feb 01 12:50 |
oiaohm2 | You really should have one just in cases these problems happen. | Feb 01 12:50 |
oiaohm2 | Its lot faster to upload the backups and check them into local than to remote. | Feb 01 12:51 |
schestowitz | I thought to meant for hosting the site with remote mysql on my pc | Feb 01 12:52 |
oiaohm2 | I was meaning for testing of backups. | Feb 01 12:52 |
schestowitz | Oh | Feb 01 12:53 |
oiaohm2 | Nothing is worse than thinking you have a backup when you don't. | Feb 01 12:53 |
schestowitz | Yes | Feb 01 12:53 |
schestowitz | The host has backups too, but not as frequent, AKAIK | Feb 01 12:53 |
schestowitz | AFAIK | Feb 01 12:53 |
oiaohm2 | We basically need to know if todays backup is complete compared to yesterdays. | Feb 01 12:54 |
oiaohm2 | Ie if current problem is a upload failure or if its a bad backup. | Feb 01 12:54 |
schestowitz | Yes, I agree | Feb 01 12:54 |
schestowitz | I look at my backups | Feb 01 12:55 |
oiaohm2 | Best way for that is loading the both into the local mysql to make sure they restore. | Feb 01 12:55 |
schestowitz | They grew every week. | Feb 01 12:55 |
schestowitz | 13mb.. 14mb.. 18 mb.. 20mb | Feb 01 12:55 |
oiaohm2 | Grew is not a good test. | Feb 01 12:55 |
oiaohm2 | They can grow yet not be restorable. | Feb 01 12:55 |
oiaohm2 | Restored is always the test. | Feb 01 12:55 |
oiaohm2 | Backup process. Backup from server. Restore on to local check local if fine backup good. | Feb 01 12:56 |
schestowitz | I think that because the upload is done on Firefox and the page grows massive, then maybe it times out with incomplete DB upload | Feb 01 12:56 |
schestowitz | oiaohm2: I'll use this tip from now on :-) | Feb 01 12:57 |
oiaohm2 | If we knew the backup was good we could be sure of that. | Feb 01 12:57 |
oiaohm2 | Also after restore is proved checksum the backup. | Feb 01 12:57 |
oiaohm2 | So any other damage can be detected. | Feb 01 12:57 |
oiaohm2 | Yes the process of backing up a database is longer than most people think. | Feb 01 12:57 |
oiaohm2 | We need to do the other parts of the backup process we we know where we stand. | Feb 01 12:58 |
schestowitz | Yes | Feb 01 12:58 |
schestowitz | I sort of trusted it cause it grew | Feb 01 12:58 |
schestowitz | I still suspsect it's tyhe browser thing | Feb 01 12:58 |
oiaohm2 | Thing is the page is partly up at the moment. | Feb 01 12:59 |
oiaohm2 | Partly is better than nothing. | Feb 01 12:59 |
schestowitz | When the page grows past some threshold, then the Web-based front-end might just.. | Feb 01 12:59 |
schestowitz | oiaohm2: it's up cause WP connects to the DB and then starts rendering. | Feb 01 12:59 |
oiaohm2 | If its the web based front end sending another could make it worse. | Feb 01 12:59 |
oiaohm2 | Ie no page rendering at all. | Feb 01 12:59 |
oiaohm2 | Because it will all depend on where it breaking. | Feb 01 13:00 |
oiaohm2 | If its the backup we can send another. | Feb 01 13:00 |
oiaohm2 | If it rains it pores on you today schestowitz | Feb 01 13:00 |
schestowitz | That's OK | Feb 01 13:00 |
schestowitz | we have other backups (host, home dir | Feb 01 13:01 |
schestowitz | Being stressed won't help, will it | Feb 01 13:01 |
schestowitz | OK | Feb 01 13:01 |
schestowitz | Wait... | Feb 01 13:01 |
schestowitz | Aha | Feb 01 13:01 |
oiaohm2 | I have mastered the art of not being stressed. | Feb 01 13:01 |
schestowitz | Wait.. | Feb 01 13:01 |
schestowitz | I look at my backup from last month.. | Feb 01 13:02 |
schestowitz | Inside thee gzip I have wrdp1.sql | Feb 01 13:02 |
schestowitz | Sice is 71MB | Feb 01 13:02 |
schestowitz | It's possible that we did another 5-10 MB in the past month | Feb 01 13:02 |
schestowitz | Does that make sense? | Feb 01 13:03 |
oiaohm2 | Not really sql is text based. | Feb 01 13:03 |
oiaohm2 | mysql internal storage is binary. | Feb 01 13:03 |
schestowitz | But regarding size | Feb 01 13:03 |
oiaohm2 | So sql gives no size compare. | Feb 01 13:03 |
oiaohm2 | Ok what is current backups .sql size. | Feb 01 13:04 |
schestowitz | Does this mean that restoring to about 40MB means we don't pass the whole thing over the Web? | Feb 01 13:04 |
schestowitz | The lastest one (this morning , 7AM) is 82MB | Feb 01 13:04 |
oiaohm2 | No binary vs .sql binary storage is smaller. | Feb 01 13:04 |
schestowitz | That's the SQL gile | Feb 01 13:05 |
schestowitz | *file | Feb 01 13:05 |
oiaohm2 | Its the reason why you need to restore those .sql into a local server. | Feb 01 13:05 |
oiaohm2 | So you can see what there size should be. | Feb 01 13:05 |
schestowitz | But wait... | Feb 01 13:05 |
schestowitz | Does that make sense? | Feb 01 13:05 |
schestowitz | Is it true then that the restored DB ought to be 80+MB if restored properly? | Feb 01 13:06 |
oiaohm2 | It might be right too. | Feb 01 13:06 |
oiaohm2 | Depends how much in the .sql struct stores smaller when converted to mysql internal binary format. | Feb 01 13:06 |
schestowitz | I've just attempted another resotre | Feb 01 13:07 |
schestowitz | Let me check... | Feb 01 13:07 |
oiaohm2 | Basically you have to restore into a mysql server to size. | Feb 01 13:07 |
schestowitz | Oh, I see... | Feb 01 13:07 |
schestowitz | OK GOOD | Feb 01 13:10 |
schestowitz | Now my restore is 47MB in size | Feb 01 13:10 |
schestowitz | Compared to 37MB beforehand | Feb 01 13:10 |
schestowitz | This means that the same file gave inconsistent resored due to browser issue, IMHO | Feb 01 13:11 |
schestowitz | *sigh of relief* | Feb 01 13:12 |
schestowitz | Maybe it's still uploading it actually | Feb 01 13:12 |
schestowitz | Yes | Feb 01 13:13 |
schestowitz | I can see now... it doesn't restore fully. | Feb 01 13:14 |
schestowitz | I wish it didn't ECHO everything that's restored | Feb 01 13:14 |
oiaohm2 | Great | Feb 01 13:14 |
schestowitz | So the DB is in okay shape | Feb 01 13:15 |
oiaohm2 | Would have been simpler if we could 100 percent trusted the backups. | Feb 01 13:15 |
schestowitz | I think the old one too is OK | Feb 01 13:15 |
schestowitz | But it has no user attached | Feb 01 13:15 |
oiaohm2 | I am not sure. | Feb 01 13:15 |
oiaohm2 | Its strange the user disappeared. | Feb 01 13:15 |
schestowitz | Then again | Feb 01 13:15 |
oiaohm2 | I still want host side backup before you play with that one. | Feb 01 13:15 |
schestowitz | I think WP would complain about user not being there, not suggest that I 'install' WP | Feb 01 13:16 |
schestowitz | oiaohm2: thank you SO much | Feb 01 13:16 |
oiaohm2 | I have give a lot of good notes make sure you reread this log and build your self some good producures. | Feb 01 13:17 |
oiaohm2 | Makes if it happens again lot less worring. | Feb 01 13:17 |
schestowitz | Mostly about testing BUs locally | Feb 01 13:18 |
schestowitz | And cases in which this may recur | Feb 01 13:18 |
schestowitz | We had an incident in 2006 or 2007 | Feb 01 13:18 |
schestowitz | Shane handled it | Feb 01 13:18 |
oiaohm2 | You need issue handling on paper. | Feb 01 13:19 |
oiaohm2 | Like not altering failed one without it backed up on host. | Feb 01 13:19 |
oiaohm2 | If you take my notes here you can build yourself handling instructions just to follow in case of failure. | Feb 01 13:19 |
oiaohm2 | Lot less worring. | Feb 01 13:20 |
schestowitz | I suppose you do some of this for a living (sysadmining) | Feb 01 13:20 |
*Omar871 (n=omar@94.249.1.158) has left #boycottnovell | Feb 01 13:20 |
oiaohm2 | I am not what you call a normal system admin. | Feb 01 13:21 |
schestowitz | Yes, the upload is cut abruptly. I can see it now | Feb 01 13:21 |
oiaohm2 | I am data recovery and system admin. | Feb 01 13:21 |
schestowitz | But it depends on how much load.. | Feb 01 13:21 |
schestowitz | I guess the browser times out or something | Feb 01 13:21 |
schestowitz | 8-10-09 03:00:59','',0,'http://boycottnovell.com/2008/10/08/r...',0,'' ,'',0)-------------- | Feb 01 13:21 |
schestowitz | END of ECHO | Feb 01 13:21 |
schestowitz | No closing brackets even | Feb 01 13:22 |
schestowitz | And in the previous one it didn't get this far | Feb 01 13:22 |
schestowitz | 37MB vs 40 seomthing | Feb 01 13:22 |
schestowitz | Maybe I should try another Web browser | Feb 01 13:22 |
oiaohm2 | That is kinda the next stop. | Feb 01 13:23 |
oiaohm2 | Worst case is that is a pure time out limit. | Feb 01 13:23 |
schestowitz | I do it from Konqueror now | Feb 01 13:24 |
schestowitz | oiaohm2: that's what I reckon | Feb 01 13:24 |
schestowitz | The first time Firefox didn't even turn out a favicon, which I though was wierd | Feb 01 13:25 |
schestowitz | *weird | Feb 01 13:25 |
schestowitz | This time it was 40 megs. Reattempting. | Feb 01 13:31 |
schestowitz | It sort of doesn't complete the transaction | Feb 01 13:32 |
*schestowitz uploading again (letting the resources be available to the browser) | Feb 01 13:36 |
*oiaohm3 (n=oiaohm@240.120.dsl.brs.iprimus.net.au) has joined #boycottnovell | Feb 01 13:40 |
oiaohm3 | Bugger connction died | Feb 01 13:40 |
oiaohm3 | Try Konqueror really what do you have to lose. schestowitz | Feb 01 13:41 |
schestowitz | http://boycottnovell.com/irc-channel/ | Feb 01 13:41 |
oiaohm3 | I really do need to find my irc password. | Feb 01 13:41 |
schestowitz | Oops. Wrong windows | Feb 01 13:41 |
*lar2 (wonderland) has joined #boycottnovell | Feb 01 13:41 |
schestowitz | oiaohm3: I do pw now | Feb 01 13:42 |
lar2 | ok figured out IRC | Feb 01 13:42 |
schestowitz | Hey, lar2 | Feb 01 13:42 |
lar2 | hey | Feb 01 13:42 |
schestowitz | We had a DB failure | Feb 01 13:42 |
lar2 | what symptoms | Feb 01 13:42 |
schestowitz | I suppose you saw the error | Feb 01 13:42 |
lar2 | ? | Feb 01 13:42 |
lar2 | yes | Feb 01 13:42 |
lar2 | wp | Feb 01 13:42 |
schestowitz | oiaohm2 here is a DB recovery expert we're lucky to have | Feb 01 13:42 |
schestowitz | mysql, not wp | Feb 01 13:43 |
schestowitz | Yes, wp couldn't find the DB or user | Feb 01 13:43 |
schestowitz | WordPress wouldn't tell | Feb 01 13:43 |
oiaohm3 | Currnet oiaohm3 until oiaohm2 dies | Feb 01 13:43 |
schestowitz | But the db vanished from the backup page in cpanel | Feb 01 13:43 |
schestowitz | But it's still there in DB list | Feb 01 13:43 |
oiaohm3 | Yes I really do need to find where I put that password for my IRC account. | Feb 01 13:43 |
lar2 | was the wp user still there in the table 'mysql' ? | Feb 01 13:43 |
schestowitz | I can't access it from phpmysqladmin | Feb 01 13:43 |
schestowitz | I can still see it there though | Feb 01 13:43 |
schestowitz | It also lost attachment to the user | Feb 01 13:44 |
schestowitz | DB user that is | Feb 01 13:44 |
lar2 | how about accessing it from the client in the shell | Feb 01 13:44 |
oiaohm3 | Remote host lar2 | Feb 01 13:44 |
schestowitz | So what you see in bn.com now is a sem-complete restore | Feb 01 13:44 |
oiaohm3 | Makes life a little hard. | Feb 01 13:44 |
schestowitz | I can't get it to restore the whole lump | Feb 01 13:44 |
lar2 | if it's really hosed you need to run from mysqladmin | Feb 01 13:44 |
schestowitz | But I have full BU | Feb 01 13:44 |
lar2 | you might have it set up to make incremental backups every time you post an article | Feb 01 13:44 |
schestowitz | lar2: user as in user table or user as in DB user? | Feb 01 13:45 |
lar2 | user as in the table 'mysql' | Feb 01 13:45 |
lar2 | that's the table with all the db users | Feb 01 13:45 |
schestowitz | lar2: yes | Feb 01 13:45 |
lar2 | that points to privileges to the database that wp uses | Feb 01 13:45 |
oiaohm3 | Issue here not all users disappeared. | Feb 01 13:45 |
schestowitz | The backup you see now in BN was not /restored/ fully | Feb 01 13:45 |
schestowitz | But I have pretty good backup from 7am | Feb 01 13:45 |
oiaohm3 | Ie db users. | Feb 01 13:46 |
schestowitz | The rest I could get from static front page or restore manually, but I 'd rather wait for host to check original DB | Feb 01 13:46 |
schestowitz | The one that's 'broken' | Feb 01 13:46 |
lar2 | is there an external (ideally stackless) log server to do forensics with ? | Feb 01 13:46 |
schestowitz | I can't even "check" or "repair" ols DB. It's there, but cpanel doesn't let me play with it | Feb 01 13:47 |
lar2 | can you get access to a shell and run 'mysql' right from the shell? | Feb 01 13:47 |
schestowitz | oiaohm2: wiki users is still there for wiki DB | Feb 01 13:47 |
schestowitz | lar2: log server for mysql? | Feb 01 13:48 |
oiaohm3 | Exactly mysql did not complete drop the ball. | Feb 01 13:48 |
lar2 | yes, logserver for mysql, apache and the host | Feb 01 13:48 |
schestowitz | I only have the web log (mostly php errors) | Feb 01 13:48 |
schestowitz | I looked at it | Feb 01 13:48 |
schestowitz | Nothing interesting there | Feb 01 13:48 |
schestowitz | Just some old php warnings. | Feb 01 13:48 |
schestowitz | I only have access to http log | Feb 01 13:49 |
schestowitz | maybe ftp too | Feb 01 13:49 |
lar2 | web interfaces for administration only work when everything is perfect, so the main administrators should always strive to stay in practice with the shell | Feb 01 13:49 |
schestowitz | But not anything beneath it | Feb 01 13:49 |
lar2 | web log might be enough, but there should also be a mysql log somewhere | Feb 01 13:49 |
schestowitz | What filename? | Feb 01 13:49 |
*oiaohm2 has quit (Read error: 110 (Connection timed out)) | Feb 01 13:50 |
oiaohm3 | Paided hosting lar2. shell not alway forth coming. | Feb 01 13:50 |
lar2 | just a sec, gotta install mysql-server | Feb 01 13:50 |
schestowitz | Thanks a lot | Feb 01 13:50 |
lar2 | oiaohm3: no shell, no money from me! | Feb 01 13:51 |
lar2 | shell is a prerequisite for hosting@! | Feb 01 13:51 |
oiaohm3 | This case we have to work with what we have lar2 no what we wish for. | Feb 01 13:52 |
oiaohm3 | I asked if he had shell as one of the first things. | Feb 01 13:52 |
lar2 | yes, the cleanup must occur first , then move to shell account | Feb 01 13:53 |
lar2 | slow net : 216kB/s | Feb 01 13:53 |
lar2 | mysql logs are often here: | Feb 01 13:54 |
lar2 | $ find /var/log/ -name 'mysql*' -print | Feb 01 13:54 |
lar2 | /var/log/mysql.log | Feb 01 13:54 |
lar2 | /var/log/mysql.err | Feb 01 13:54 |
lar2 | /var/log/mysql | Feb 01 13:54 |
schestowitz | Oh | Feb 01 13:54 |
schestowitz | I can't get that 'far' | Feb 01 13:54 |
oiaohm3 | We are very much down to host restore. | Feb 01 13:54 |
schestowitz | unless I do a cronjob maybe | Feb 01 13:54 |
oiaohm3 | maybe is the real thing here. | Feb 01 13:55 |
oiaohm3 | Secuirty can block your access from that as well. | Feb 01 13:55 |
schestowitz | Yes, I know. | Feb 01 13:55 |
oiaohm3 | I have to get to bed is midnight here. | Feb 01 13:55 |
schestowitz | oiaohm3: thank you for the help | Feb 01 13:55 |
oiaohm3 | More sane time where you are lar2? | Feb 01 13:55 |
schestowitz | lar2: I think I'll try to restore from morning backup (7am) and then ask host to check the old DB (11am) | Feb 01 13:56 |
schestowitz | In the mean time I created a second similar DB to just serve pages. | Feb 01 13:57 |
schestowitz | But I can't upload it properly. | Feb 01 13:57 |
oiaohm3 | I would recommend db3 | Feb 01 13:57 |
schestowitz | It seems to time out at the browser, but at varying points along the way depending on timeouts, system load.. | Feb 01 13:57 |
schestowitz | A third one? | Feb 01 13:57 |
oiaohm3 | Ie third one for upload attempts | Feb 01 13:57 |
schestowitz | Oh | Feb 01 13:58 |
oiaohm3 | So you only stay on the best. | Feb 01 13:58 |
schestowitz | OK, good. | Feb 01 13:58 |
oiaohm3 | No point ended up worse. | Feb 01 13:58 |
schestowitz | Last upload was lame.. 20.53 MB | Feb 01 13:58 |
schestowitz | So I shaved off data! | Feb 01 13:58 |
schestowitz | I'll try again | Feb 01 13:58 |
oiaohm3 | If everything goes ok everything should be back to normal ish in 24 hours. | Feb 01 13:59 |
oiaohm3 | Worst case is when you end up having to email backup to host so they can restore it. | Feb 01 13:59 |
oiaohm3 | Boy that suxs | Feb 01 14:00 |
schestowitz | Yes | Feb 01 14:00 |
schestowitz | Better than rotten backups | Feb 01 14:00 |
oiaohm3 | Depends | Feb 01 14:00 |
oiaohm3 | If you internet connection has good completely suspect is hell. | Feb 01 14:01 |
oiaohm3 | good/gone | Feb 01 14:01 |
schestowitz | What do you mean? | Feb 01 14:01 |
schestowitz | Oh, well.. on the servers,, right? | Feb 01 14:01 |
schestowitz | I could work from my office if the Net was down at home. | Feb 01 14:01 |
schestowitz | Done this in Sept. | Feb 01 14:02 |
oiaohm3 | I have had the busness adsl line here go problem. | Feb 01 14:02 |
oiaohm3 | Without notice. So tried to restore threw web failed. Tried to email it up and it was also damaged. | Feb 01 14:02 |
oiaohm3 | Reason line I had was stuffed. | Feb 01 14:02 |
oiaohm3 | Took ages to find out that was the case. | Feb 01 14:02 |
oiaohm3 | Your ISP static IP where you are uploading from. | Feb 01 14:03 |
oiaohm3 | Just had a bad idea have had problems from time to time with large uploads from dynamic IP ISP's. | Feb 01 14:03 |
oiaohm3 | Hopefully by the time I am back on Line you still don't need me schestowitz | Feb 01 14:04 |
*oiaohm3 has quit (Remote closed the connection) | Feb 01 14:04 |
schestowitz | lar2: I'm starting a third DB | Feb 01 14:07 |
schestowitz | I wish cpanel wasn't stupid enough to ECHO the whole thing doing restore | Feb 01 14:07 |
trmanco | schestowitz, another fish nym on COLA? | Feb 01 14:09 |
schestowitz | Yes, many appear all the time | Feb 01 14:09 |
trmanco | so BN is infected with a virus :-P | Feb 01 14:10 |
schestowitz | No | Feb 01 14:10 |
schestowitz | Host errors | Feb 01 14:10 |
schestowitz | yesterday it was too much load | Feb 01 14:10 |
schestowitz | And today the DB went bonkers | Feb 01 14:10 |
schestowitz | Related to load too | Feb 01 14:10 |
schestowitz | So it needs attention from the host | Feb 01 14:11 |
schestowitz | It ought to be possible to fix, but I don't want to do anything without them backing it up first | Feb 01 14:11 |
lar2 | just a complex thought here | Feb 01 14:11 |
lar2 | mysql can do automatic replication, that could be used to mirror the data off site | Feb 01 14:11 |
lar2 | from there full backups could be done each time a new article is added with less load on the main server | Feb 01 14:11 |
schestowitz | Really? | Feb 01 14:11 |
trmanco | schestowitz, I was just kidding :-O | Feb 01 14:11 |
trmanco | :-P* | Feb 01 14:11 |
schestowitz | Can I replicate without CLI access? | Feb 01 14:11 |
schestowitz | trmanco: I know | Feb 01 14:11 |
lar2 | CLI ? | Feb 01 14:11 |
schestowitz | trmanco: but it's easy for some people to be fooled. | Feb 01 14:12 |
trmanco | they should just check from themselves | Feb 01 14:12 |
schestowitz | lar2: how can I replicate a DB? | Feb 01 14:12 |
schestowitz | in Cpanel that is.. | Feb 01 14:12 |
schestowitz | I have: | Feb 01 14:13 |
schestowitz | boycottn_wiki 2.05 MB | Feb 01 14:13 |
schestowitz | boycottn_wki Delete User | Feb 01 14:13 |
schestowitz | Delete Database | Feb 01 14:13 |
schestowitz | boycottn_wrdp1 81.41 MB | Feb 01 14:13 |
schestowitz | Delete Database | Feb 01 14:13 |
schestowitz | boycottn_wrdp2 45.35 MB | Feb 01 14:13 |
schestowitz | boycottn_wrdp2 Delete User | Feb 01 14:13 |
schestowitz | Delete Database | Feb 01 14:13 |
schestowitz | boycottn_wrdp3 45.35 MB | Feb 01 14:13 |
schestowitz | Delete Database | Feb 01 14:13 |
trmanco | :O | Feb 01 14:13 |
trmanco | 3 databases? | Feb 01 14:13 |
schestowitz | That second and third (well, 3rd and 4th) are my uploads | Feb 01 14:14 |
schestowitz | trmanco: the first is the original | Feb 01 14:14 |
schestowitz | I needs admins to check it | Feb 01 14:14 |
trmanco | wow 81 megs is huge | Feb 01 14:14 |
schestowitz | I try to replicate onto other dbs and then load BN from these | Feb 01 14:14 |
trmanco | mine only gas aout 7 :-P | Feb 01 14:14 |
schestowitz | trmanco: yes, we have a lot of text | Feb 01 14:14 |
trmanco | you can make a landing page, with some mirrors | Feb 01 14:15 |
lar2 | for replication, you'll need at least on other machine, any *nix will do, even your desktop | Feb 01 14:15 |
schestowitz | If only I could copy from boycottn_wrdp1 to anoter slot | Feb 01 14:15 |
trmanco | schestowitz, 404 no found? | Feb 01 14:15 |
lar2 | http://dev.mysql.com/doc/refman/5... | Feb 01 14:15 |
schestowitz | trmanco: ignore the site's output for now | Feb 01 14:15 |
trmanco | ok | Feb 01 14:16 |
schestowitz | I just try to restore something temporarily | Feb 01 14:16 |
trmanco | schestowitz, if you need some help, just ask | Feb 01 14:16 |
schestowitz | Thanks. | Feb 01 14:16 |
schestowitz | lar2: do you suggest I ran the DB locally? | Feb 01 14:17 |
trmanco | my host can try and handle some of your load | Feb 01 14:17 |
*schestowitz installed mysql | Feb 01 14:18 |
schestowitz | I'll need it to test backups anyway | Feb 01 14:18 |
lar2 | yes, that's one way run the slave locally | Feb 01 14:19 |
lar2 | it's a small db, right? <= 100MB | Feb 01 14:19 |
schestowitz | Yes | Feb 01 14:20 |
schestowitz | But heavy load, I think | Feb 01 14:20 |
schestowitz | None for Mandriva: http://dev.mysql.com/downloads... | Feb 01 14:21 |
schestowitz | The repo doesn't give the option either. | Feb 01 14:21 |
trmanco | just found this: http://identi.ca/rms | Feb 01 14:22 |
schestowitz | OK, I contacted the host | Feb 01 14:28 |
schestowitz | Urgent - DB issue For no reason (other than heavy load), the WordPress DB, boycottn_wrdp1, was suddenly made unavailble. I can't touch it from mysqladmin and I can't make a backup of it, but it's still there on the server (I can see it in list of databases) and for some reason it also dropped its only DB-user. | Feb 01 14:28 |
schestowitz | An admin with full access to the box needs to check what happened to this DB and restore its working order with the DB-user connected to it so that WordPress can communicate with it. | Feb 01 14:28 |
schestowitz | In the mean time, I have created incomplete copies of this database (backed up hours before it sort of crashed) in boycottn_wrdp2 and boycottn_wrdp3 and I try to have WordPress access them as temporary solutions. | Feb 01 14:28 |
schestowitz | That pretty much sums up the situation | Feb 01 14:28 |
lar2 | it'd be important to find out how the only DB-user got dropped | Feb 01 14:30 |
schestowitz | Yes | Feb 01 14:31 |
schestowitz | There is another user in Wiki DB | Feb 01 14:31 |
schestowitz | It stayed... and so has the DB | Feb 01 14:31 |
schestowitz | So the DB decided to run away (it's still there, just not accessible) and the user is not tied to it. | Feb 01 14:32 |
schestowitz | Linux injured KDE: http://aseigo.blogspot.com/2009/01/i-... | Feb 01 14:33 |
lar2 | the user from the mysql db or from the boycottn_wrdp db ? | Feb 01 14:35 |
*maxstirner (n=user@xdsl-87-79-134-116.netcologne.de) has joined #boycottnovell | Feb 01 14:35 |
schestowitz | From the mysql db | Feb 01 14:35 |
schestowitz | Not the WordPress one. | Feb 01 14:35 |
maxstirner | oh youre discussing the problem already | Feb 01 14:36 |
maxstirner | was just tgoing to point out.. | Feb 01 14:36 |
schestowitz | Yes.. | Feb 01 14:36 |
schestowitz | I get E-mails about it too. | Feb 01 14:36 |
schestowitz | Konqueror stops uploading after 45 MB | Feb 01 14:37 |
schestowitz | maxstirner: I try to restore from backup to new slots | Feb 01 14:37 |
schestowitz | But it's too big for the browsers | Feb 01 14:37 |
lar2 | scp, sftp instead | Feb 01 14:37 |
lar2 | or use ssh to tunnel to the mysql db and use the clients locally | Feb 01 14:38 |
schestowitz | lar2: where need to mysql file be though? | Feb 01 14:38 |
schestowitz | Not in public_html or /home/useraccount | Feb 01 14:39 |
lar2 | for the ssh tunnel, it can be on your local machine | Feb 01 14:39 |
schestowitz | You're thinking about running the site off my desktop | Feb 01 14:39 |
lar2 | not in that example, | Feb 01 14:39 |
schestowitz | Oh | Feb 01 14:40 |
lar2 | that would be restoring the server from your desktop | Feb 01 14:40 |
schestowitz | But I can't SSH to my server | Feb 01 14:40 |
lar2 | ?!?!?!?! | Feb 01 14:40 |
schestowitz | I can't | Feb 01 14:40 |
lar2 | what happened | Feb 01 14:40 |
schestowitz | Host won't let that be | Feb 01 14:40 |
lar2 | new host! | Feb 01 14:40 |
*maxstirner (n=user@xdsl-87-79-134-116.netcologne.de) has left #boycottnovell | Feb 01 14:40 |
schestowitz | I can't bear the thought of E-mailing people big mysql DBs | Feb 01 14:42 |
schestowitz | With passwords and all. | Feb 01 14:42 |
schestowitz | Konqueror uploads just 45.35 MB, FF does 20.53 MB | Feb 01 14:43 |
lar2 | don't email data, please, please, that's an artifact of how the technological failure of MS-based infrastructure have damaged behaviors | Feb 01 14:43 |
schestowitz | Let me try to upload from the Fedora rooms | Feb 01 14:44 |
schestowitz | *LOL* | Feb 01 14:46 |
schestowitz | Firefox 3 gives me EULA on Feodra | Feb 01 14:46 |
schestowitz | let's try Opera | Feb 01 14:55 |
schestowitz | Opera does decently so far | Feb 01 14:59 |
schestowitz | Let's see when it gives up.. | Feb 01 14:59 |
schestowitz | Firefox give up at the earliest stage, Konqueror better, but the testing conditions are not identical. | Feb 01 15:00 |
lar2 | you are uploading to the host service via some upload form? | Feb 01 15:00 |
schestowitz | Yes | Feb 01 15:00 |
schestowitz | Precisely | Feb 01 15:00 |
schestowitz | I use the Fedora clusters now | Feb 01 15:00 |
schestowitz | More powerful machines, tons of software installed on the domain | Feb 01 15:01 |
schestowitz | Opera still going :-) | Feb 01 15:01 |
schestowitz | 48 MB | Feb 01 15:01 |
schestowitz | It needs to reach 81 | Feb 01 15:01 |
schestowitz | Firefox does about 21 on both Mandriva and Fedora | Feb 01 15:01 |
schestowitz | Sometimes 40MB | Feb 01 15:01 |
schestowitz | Firefox 2 and 3 respectively | Feb 01 15:01 |
schestowitz | 57MB :-) | Feb 01 15:02 |
schestowitz | If it completes it, then BN become accessible at least | Feb 01 15:02 |
schestowitz | We lost another Digg front page because of this mess | Feb 01 15:02 |
schestowitz | 65MB.. | Feb 01 15:02 |
schestowitz | Fingers crossed.. | Feb 01 15:02 |
schestowitz | Ah! Bless Norway, now let's change some vars. | Feb 01 15:04 |
schestowitz | boycottn_wrdp1 81.41 MB | Feb 01 15:04 |
schestowitz | boycottn_wrdp3 80.89 MB | Feb 01 15:04 |
schestowitz | 4-hour diff | Feb 01 15:04 |
schestowitz | OK :-) | Feb 01 15:05 |
schestowitz | We didn't lose much then. | Feb 01 15:05 |
schestowitz | Maybe it's better to just let the old DB done and restore the diffs manually. | Feb 01 15:05 |
schestowitz | http://boycottnovell.com/ | Feb 01 15:06 |
schestowitz | Bless Opera! | Feb 01 15:06 |
lar2 | excellent | Feb 01 15:07 |
schestowitz | Hold on | Feb 01 15:11 |
schestowitz | I'll ensure no data loss | Feb 01 15:11 |
lar2 | ok but now the forensics to be able to avoid a repeat | Feb 01 15:16 |
schestowitz | OK comments shut | Feb 01 15:17 |
schestowitz | I posted a notificartion | Feb 01 15:17 |
schestowitz | Yes, that's the thing | Feb 01 15:17 |
schestowitz | I'd rather use the old DB after diagnostics | Feb 01 15:17 |
schestowitz | This one at least serves as an archive | Feb 01 15:18 |
schestowitz | Most of our visitors (like 50%+) are not subscribers but people who read our archives (5000 posts) | Feb 01 15:18 |
schestowitz | So I don't mind freezing it like Groklaw for a little while. | Feb 01 15:18 |
schestowitz | We lost a front page methinks :-( ... http://digg.com/linux_unix/Bill_Gates_on_Lin... | Feb 01 15:19 |
lar2 | too bad, | Feb 01 15:20 |
lar2 | it's a good one though | Feb 01 15:20 |
lar2 | the iowa case had good stuff online | Feb 01 15:21 |
schestowitz | Yeah.. | Feb 01 15:21 |
schestowitz | But I began worrying about it alll... you know, being resotred | Feb 01 15:21 |
schestowitz | PJ mailed me last nigth | Feb 01 15:21 |
lar2 | which is probably why ms settled so quickly | Feb 01 15:21 |
schestowitz | She's not posting properly anymore, just member-onlies. | Feb 01 15:21 |
lar2 | i noticed | Feb 01 15:21 |
schestowitz | With "sign --Groklaw team") | Feb 01 15:21 |
schestowitz | *sined | Feb 01 15:21 |
schestowitz | *signed | Feb 01 15:21 |
lar2 | have they stepped up the harassment or threats against her? | Feb 01 15:22 |
schestowitz | But the preservation part if importanyt | Feb 01 15:22 |
schestowitz | We get heaps of archives being used and not by the chair (to which we preach) | Feb 01 15:22 |
schestowitz | lar2: if so, they got their way, which pisses me off | Feb 01 15:22 |
lar2 | the preservation is very useful, but they could be better indexed or cataloged | Feb 01 15:22 |
schestowitz | When I was a PhD student (2006) I was getting lots of harrassment like that for MERELY PROMOTING Linux | Feb 01 15:23 |
schestowitz | So I said, look... if I quit now, they win | Feb 01 15:23 |
schestowitz | Instead they only urged me to dig up the filth and write about my findings | Feb 01 15:23 |
schestowitz | lar2: indexing will be done over time. When I have many pages like OOXML corruption, then I set up an index | Feb 01 15:24 |
schestowitz | Groklaw rebuilds SCO indices now. | Feb 01 15:24 |
schestowitz | Because SCO is practically over if you ignore yesterday noise from MOG | Feb 01 15:24 |
lar2 | yes, but she seems to be missing or not willing to deal with the appearance that everything with SCO the last 3 years seems to be setting the stage for novell to be the next sco | Feb 01 15:32 |
schestowitz | Yes | Feb 01 15:33 |
schestowitz | Someone else said this | Feb 01 15:33 |
schestowitz | I believe she's concerned about posting SCO-hostile stuff. I don't know if the threats keep it quiet. Either way, that along with Technocrat and Linux.com, is a major loss to us as a reporting asset | Feb 01 15:33 |
lar2 | i have some comments offline about technocrat, | Feb 01 15:39 |
lar2 | i saw newsforge folding *way* back, then linux.com | Feb 01 15:39 |
lar2 | it'd be useful to get daylight on what's going on with linux.com | Feb 01 15:39 |
schestowitz | Yes | Feb 01 15:42 |
schestowitz | I look carefully | Feb 01 15:42 |
schestowitz | Nothing major have I heard since........ | Feb 01 15:42 |
schestowitz | Technocrat.. same here. Lost comments... I mailed Bruce about it asking for archives to stay, but alas no response ... | Feb 01 15:43 |
schestowitz | China birth defects 'up sharply' < http://news.bbc.co.uk/1/hi/worl... > Pollution deaths | Feb 01 15:43 |
lar2 | yeah defect-causing pollution is very wide spread, more than most are willing to face up to... | Feb 01 15:46 |
lar2 | it's basically everywhere | Feb 01 15:46 |
schestowitz | In the UK, you now get fines for shooting pics.. http://www.flickr.com/photos/22658121@N00... (The €£50 Photograph) | Feb 01 15:47 |
lar2 | even in norway, the brits have dump nuke cores which make there way quickly up the food chain till the birds shit technium onto the land | Feb 01 15:47 |
schestowitz | "Then a voice behind me demanded that I stop whatever I was doing. It was a policeman. Two policemen. They talked in very stern terms about terrorism and security.. But is this kind of incident getting out of control? I'm hearing more & more stories like this from friends who take photographs, and quite frankly, it scares the hell out of me. " | Feb 01 15:47 |
lar2 | lame that the photographers knuckled under | Feb 01 15:48 |
schestowitz | I just LOVE this one: http://www.boingboing.net/2008/07/... | Feb 01 15:49 |
schestowitz | They let the white-collar criminals go and chase old ladies that are supposedly child-molesting terrorists because they bought some cameras they play with on a trip | Feb 01 15:49 |
lar2 | reminds me of a story a friend had when pumping gas in SC one Dec 23rd, | Feb 01 15:50 |
lar2 | the sheriff gave him the option of paying the fine right then and there, in cash, or settin his as in jail till the sheriff's brother inlaw the judge could come back after newyears to hear the case... | Feb 01 15:50 |
lar2 | hey if they were using the laws for pursuing white collar crime pretty much every MS gold partner would be locked up, it's much easier and for some small-balled individuals to go shake down old ladies instead | Feb 01 15:51 |
schestowitz | In a world of Barbarians, punching someone in the face is acutely "impolite" | Feb 01 15:51 |
schestowitz | “Microsoft’s perspective is best reflected by Bob Herbold, Chief Operating Officer, to whom the CFO reports. Bob very sincerely replied, “Bill, everyone is doing it.”" http://www.billparish.com/msftfra... | Feb 01 15:52 |
lar2 | they appear to have copied enrons methods after running the loss in 1998 | Feb 01 15:53 |
schestowitz | Sony's Profits Down 95% < http://www.osnews.com/story/20889/Son... > | Feb 01 15:53 |
schestowitz | Cost of DRM, eh? | Feb 01 15:53 |
lar2 | wouln't surprise me if most of their money comes from printing new stock / buybacks | Feb 01 15:53 |
schestowitz | http://schestowitz.com/temp/research... | Feb 01 15:53 |
schestowitz | I'm still poised to do an article about DRM (Comes) | Feb 01 15:54 |
schestowitz | But BN is on freeze now | Feb 01 15:54 |
schestowitz | http://boycottnovell.com/wiki/ind... | Feb 01 15:54 |
schestowitz | lar2: re: printing money, see the Bill Parish piece. That's just what they do. The gains go from Mom&Pop investors (esp. in the past year) into the 'safe' (Ballmer@Switzerland, Gates Foundry) | Feb 01 15:55 |
PetoKraus | schestowitz: how's your DB restore goint? | Feb 01 15:56 |
schestowitz | Ech.. | Feb 01 15:56 |
schestowitz | I'll stop posting until the host comes up with something | Feb 01 15:57 |
PetoKraus | yeah | Feb 01 15:57 |
schestowitz | No data was lost though, so no duplicate labour | Feb 01 15:57 |
schestowitz | I could restore within 20-30 minutes, but I could just wait and hear a diagnosis | Feb 01 15:58 |
schestowitz | Merging two DBs isn't fun, not at this scale (81MB) | Feb 01 15:58 |
PetoKraus | yah | Feb 01 15:58 |
schestowitz | Microsoft 'Quality Control' = [Q]uality for the stock and for Microsoft to [C]ontrol minds: http://www.osnews.com/story/20884/Only_O... The [U]ser?? Na... that's just a 'side-effect... let them east Vista | Feb 01 15:59 |
lar2 | anything about windows 7 is really just chaff - the topic at hand is vista failure, vista drm and ubuntu and other distros rocketing | Feb 01 16:00 |
schestowitz | No | Feb 01 16:00 |
schestowitz | They change the topic :-) | Feb 01 16:00 |
schestowitz | That's the thing | Feb 01 16:00 |
lar2 | only if they are allowed | Feb 01 16:00 |
schestowitz | I did a semi-scinetific thing | Feb 01 16:00 |
lar2 | elaborate pelase | Feb 01 16:00 |
schestowitz | Every Sat/Sun I look at 600 news items on 'microsoft' (Google News) | Feb 01 16:01 |
schestowitz | I handle them by slicing with filters | Feb 01 16:01 |
schestowitz | First 'vista' | Feb 01 16:01 |
schestowitz | then 'down 7' | Feb 01 16:01 |
schestowitz | *'dows 7' | Feb 01 16:01 |
schestowitz | Guess what? Vista doesn't exist anymore | Feb 01 16:01 |
lar2 | i often wonder how much help they get with that from insiders at google | Feb 01 16:01 |
schestowitz | For the past 2 months I have done this test. Wrote about this briefly in BN (twice) and also in IRC and USENET | Feb 01 16:02 |
lar2 | google has been polluting itself with MS hires for a long while and leopards don't change their spots | Feb 01 16:02 |
schestowitz | search the GOOG for "Vista 7" | Feb 01 16:02 |
schestowitz | Page 1 is us | Feb 01 16:02 |
lar2 | LOL | Feb 01 16:02 |
schestowitz | Apropos | Feb 01 16:02 |
schestowitz | Wait.. | Feb 01 16:02 |
lar2 | IIRC opendocument was torpedoed in google news starting since aug 2007 | Feb 01 16:02 |
schestowitz | OK, got it... | Feb 01 16:03 |
schestowitz | http://www.techcrunch.com/2009/01/... | Feb 01 16:03 |
schestowitz | See these | Feb 01 16:03 |
schestowitz | "From: Stephen | Feb 01 16:04 |
schestowitz | Date: Wed, 28 May 2008 13:25:07 -0700 (PDT) | Feb 01 16:04 |
schestowitz | Local: Wed, May 28 2008 2:25 pm | Feb 01 16:04 |
schestowitz | Subject: Re: So… Why’d you left, guys? I mean, seriously. | Feb 01 16:04 |
schestowitz | Actually, I hit the Send button on this before I intended to. | Feb 01 16:04 |
schestowitz | I left Microsoft to work for Google in 2005. I stayed 10 months. I | Feb 01 16:04 |
schestowitz | was demoralized. I shouldn’t have ever taken that job. I was | Feb 01 16:04 |
schestowitz | disenchanted the whole time, and yes, like you, my regret over the | Feb 01 16:04 |
schestowitz | poor bargain I’d made affected my performance." | Feb 01 16:04 |
schestowitz | -------------- | Feb 01 16:04 |
schestowitz | "From: Ted | Feb 01 16:04 |
schestowitz | Date: Wed, 28 May 2008 17:39:06 -0700 (PDT) | Feb 01 16:04 |
schestowitz | Local: Wed, May 28 2008 6:39 pm | Feb 01 16:04 |
schestowitz | Subject: Re: So… Why’d you left, guys? I mean, seriously." | Feb 01 16:04 |
schestowitz | [...] | Feb 01 16:04 |
schestowitz | "Also I was surprised that Google seemed to be proud that they didn’t | Feb 01 16:04 |
schestowitz | communicate from one interviewer to the next: at Microsoft it was a | Feb 01 16:04 |
schestowitz | good opportunity to find more appropriate interviewers, etc. if a | Feb 01 16:04 |
schestowitz | person seemed misslated. Oh well, I thought my interview and hiring | Feb 01 16:04 |
schestowitz | process was an anomaly." | Feb 01 16:04 |
schestowitz | ===== | Feb 01 16:04 |
schestowitz | TechCrunch is run by Michael Arrington (aka "spit in face") | Feb 01 16:05 |
schestowitz | http://web2.commongate.com/post/Micr... | Feb 01 16:05 |
schestowitz | "Valleywag is taking Mike Arrington, Om Malik, and John Battelle to task for allegedly doing a Federated Media version of “pay to post”. " | Feb 01 16:05 |
schestowitz | http://www.techcrunch.com/2009/01... | Feb 01 16:05 |
schestowitz | "by Michael Arrington on January 18, 2009" | Feb 01 16:06 |
schestowitz | http://www.techcrunch.com/ | Feb 01 16:06 |
schestowitz | 2 million readers (claimed) | Feb 01 16:06 |
schestowitz | Takeaway message: "Google sucks to work for, I wanna be back to Redmond" | Feb 01 16:06 |
lar2 | trolling i presume | Feb 01 16:07 |
lar2 | i have contact with a few "former" softers | Feb 01 16:07 |
schestowitz | "In 2008 Google HR set up a private Google Group to ask former employees why they left the company. We’ve been forwarded what appears to be authentic posts to the thread by a number of ex-Googlers, which we reprint below minus identifying information other than their first names." | Feb 01 16:07 |
schestowitz | Forwarded? | Feb 01 16:07 |
schestowitz | By whom? | Feb 01 16:07 |
schestowitz | How did he get that? | Feb 01 16:07 |
schestowitz | W-E? | Feb 01 16:07 |
lar2 | lame, lazy and stupid in their own way | Feb 01 16:07 |
schestowitz | Andre de La Costa du jour? | Feb 01 16:08 |
schestowitz | In USENET (earlier today) a 18-year ols Microsoft veteran posted in the Linux advocacy NG | Feb 01 16:08 |
schestowitz | He struck back at [H]omer, who wrote about Microsoft putting $$$ on GNU's table (FSF donation) | Feb 01 16:08 |
schestowitz | I called it hush-hush money. RMS et al could use the money after all. | Feb 01 16:09 |
schestowitz | The Turfers love pointing to this < http://www.fsf.org/licensing/licenses/ > page, which now contains Microsoft licences after some prodding from LWN.net from what I could once gather. FSF very reluctantly put the ms-* in ther | Feb 01 16:10 |
lar2 | hey, one of the costs that's adding to the economic hardships is the failure of windows to be able to manage files, very different from real systems | Feb 01 16:10 |
schestowitz | That's what MCSEs are for | Feb 01 16:11 |
schestowitz | A dozen per box (overstating) | Feb 01 16:11 |
schestowitz | Rex calls em "box booters" | Feb 01 16:11 |
lar2 | yeah but they're only there to burn out the user so that they give up on getting their files back | Feb 01 16:11 |
schestowitz | Of course, you could also call them Registry/malware cleaners/defraggers etc etc | Feb 01 16:12 |
schestowitz | What I did today on the BN server was caused by million of hits on one box | Feb 01 16:12 |
schestowitz | And it hasn't happened to me in like.. EVER! | Feb 01 16:12 |
lar2 | another item is the ecological and economic catastrophes in the server room, i see that it looks like every shop with servers infected with MS have worse than 1:1 service:server ratio. | Feb 01 16:12 |
schestowitz | I never needed to repair a mysql DB in 5 years of handling (now) about 20+ DBs | Feb 01 16:13 |
lar2 | at 300W - 500W a whack, that adds up | Feb 01 16:13 |
lar2 | can you give more details (sanitized if necessary) on how the service was slashdotted? | Feb 01 16:13 |
schestowitz | The MSFTers stormed all over this one: http://digg.com/linux_unix/Microsoft_o... | Feb 01 16:13 |
schestowitz | lar2: What do you mean by /.ed? | Feb 01 16:14 |
lar2 | too much user load | Feb 01 16:14 |
schestowitz | I have it from a reliable source that /. is gamed by MS hires'peripheral agents. | Feb 01 16:14 |
schestowitz | Same in Digg, naturally (but I don't have it said from Digg staff) | Feb 01 16:14 |
lar2 | i can see that, too | Feb 01 16:14 |
lar2 | i never run into the old /. staff to be sure | Feb 01 16:15 |
schestowitz | lar2: say I need to do 200 db queries per page | Feb 01 16:15 |
schestowitz | it's not static | Feb 01 16:15 |
schestowitz | I have corss-refs (done on the fly) | Feb 01 16:15 |
schestowitz | That's how I get so many so quickly | Feb 01 16:15 |
lar2 | its to the point that background checks need to cover redmond ties | Feb 01 16:15 |
schestowitz | So if we get 1000 visit per hours, that's a lot of db thrashing | Feb 01 16:15 |
schestowitz | PLUS... we cross-link.. | Feb 01 16:15 |
lar2 | do you have pound in front? | Feb 01 16:15 |
schestowitz | You know what that means... people stick around, so more page requests | Feb 01 16:16 |
lar2 | sounds like you need a reverse proxy | Feb 01 16:16 |
schestowitz | If it loads at all.. | Feb 01 16:16 |
lar2 | the db is only needed for the first load or for interactive work | Feb 01 16:16 |
lar2 | such as comments | Feb 01 16:16 |
schestowitz | Yes | Feb 01 16:16 |
schestowitz | I use Coral | Feb 01 16:16 |
schestowitz | I close comments, then throw it over there. | Feb 01 16:16 |
schestowitz | But peole depart from the page, so it's no longer static | Feb 01 16:16 |
lar2 | http://www.squid-cache.org/ | Feb 01 16:17 |
lar2 | http://www.apsis.ch/pound/ | Feb 01 16:17 |
schestowitz | I couldn't care less about whether traffic is measured on BN or some $ProxyPoxy.. as long as people manage the read the stuff | Feb 01 16:17 |
schestowitz | WordPress has memcache | Feb 01 16:17 |
schestowitz | But Ryan Boren dropped it at the time, IIRC | Feb 01 16:18 |
schestowitz | I was more closely involved in WP at the time. | Feb 01 16:18 |
schestowitz | It caused trouble and they tropped it | Feb 01 16:18 |
schestowitz | *dropped | Feb 01 16:18 |
lar2 | no need for them to reinvent the wheel, squid or pound might do enough | Feb 01 16:18 |
schestowitz | And we use the 'Debian' version (5-year LTS) of WordPress, so it lags behind and gets security patches | Feb 01 16:18 |
schestowitz | This reminds me I need to produce the monthly WordPress-Hackers stats for the mailing list | Feb 01 16:19 |
lar2 | patches are good, upgrades == suprises == :( | Feb 01 16:19 |
lar2 | lagging behind == predictable == good | Feb 01 16:19 |
lar2 | from a sysadmin perspective | Feb 01 16:19 |
*schestowitz looks at URLS | Feb 01 16:19 |
schestowitz | Yes | Feb 01 16:19 |
schestowitz | That's what I say | Feb 01 16:19 |
schestowitz | Plugins especially | Feb 01 16:19 |
schestowitz | I'm almost frantic about having to reprogram my plugins myself | Feb 01 16:20 |
schestowitz | My cross-refs ALL depend on it. | Feb 01 16:20 |
schestowitz | About 20,000+ (!!) cross-refs | Feb 01 16:20 |
lar2 | hmm interesting research material, | Feb 01 16:20 |
schestowitz | Is Squid better than the latter | Feb 01 16:21 |
schestowitz | Which one to look at first | Feb 01 16:21 |
schestowitz | Combining too is moot/impossible | Feb 01 16:21 |
schestowitz | I also have 8GB of pages on my HDD | Feb 01 16:21 |
schestowitz | Pages that I cite I keep copy of | Feb 01 16:21 |
schestowitz | (Proven useful before because Web Archive is slow to grab 'pulled' articles) | Feb 01 16:22 |