Re: [Testbed-admins] Problem updating the database after git clone from emulab-stable

["Cheng Cui" <ccui1@tigers.lsu.edu>]

Hello,

Will somebody add the debug mode in the sql update file "sitevarscheck"
and "sitevarsdiff"? The mysql database in our local site has just a
little difference on the version number, and I think the mysqldump
options generated a little more to compare the difference between Utah
stable source and our db. For example, in a temp sitevarscheck.2191
which I generated from the "sitevarscheck":

-- MySQL dump 10.11
--
-- Host: localhost    Database: tbdb
-- ------------------------------------------------------
-- Server version	5.0.45-log
/*!40103 SET @OLD_TIME_ZONE=@@TIME_ZONE */;
/*!40103 SET TIME_ZONE='+00:00' */;
/*!40014 SET @OLD_UNIQUE_CHECKS=@@UNIQUE_CHECKS, UNIQUE_CHECKS=0 */;
/*!40014 SET @OLD_FOREIGN_KEY_CHECKS=@@FOREIGN_KEY_CHECKS,
FOREIGN_KEY_CHECKS=0 */;
/*!40101 SET @OLD_SQL_MODE=@@SQL_MODE, SQL_MODE='NO_AUTO_VALUE_ON_ZERO'
*/;
/*!40111 SET @OLD_SQL_NOTES=@@SQL_NOTES, SQL_NOTES=0 */;
--
-- Dumping data for table `sitevariables`
--

INSERT INTO `sitevariables` VALUES ('general/testvar',NULL,'43','A test
variable');
INSERT INTO `sitevariables` VALUES ('web/nologins','1','0','Non-zero
value indicates that no user may log IGNORE INTO the Web Interface;
non-admin users are auto logged out.');
INSERT INTO `sitevariables` VALUES ('web/message',NULL,'','Message to
place in large lettering under the login message on the Web
Interface.');
.........

But in the Utah stable source ../sql/sitevars-create.sql:

-- MySQL dump 10.10
--
-- Host: localhost    Database: tbdb
-- ------------------------------------------------------
-- Server version	5.0.20-log

--
-- Dumping data for table `sitevariables`
--


INSERT IGNORE INTO sitevariables VALUES ('general/testvar',NULL,'43','A
test variable');
INSERT IGNORE INTO sitevariables VALUES
('web/nologins',NULL,'0','Non-zero value indicates that no user may log
into the Web Interface; non-admin users are auto logged out.');
INSERT IGNORE INTO sitevariables VALUES ('web/message',NULL,'','Message
to place in large lettering under the login message on the Web
Interface.');

***************************************************

Looks the data are all right, but the format is just pretty tiny
different. How could I avoid the "sitevarscheck"? Or is there any better
way to debug it? Thanks.

Best Regards,

Cheng Cui
. Frey Computing Services Center . Baton Rouge, LA 70803

-----Original Message-----
From: Robert P Ricci [mailto:ricci@cs.utah.edu] 
Sent: Thursday, March 25, 2010 2:38 AM
To: Cheng Cui
Cc: 'Mike Hibler'; testbed-admins@flux.utah.edu; 'Leigh Stoller'
Subject: Re: [Testbed-admins] Problem updating the database after git
clone from emulab-stable

Try running dbfillcheck by hand with the -d flag - this will leave
around the file showing the differences, and tell you where it is.

Thus spake Cheng Cui on Wed, Mar 24, 2010 at 07:12:48PM -0500:
> It looks not the problem of sql statement. I commented out all the sql
> statements in src/../sql/database-fill.sql, but the gmake boss-install
> still gets the same error. I am sure I don't need to update my
database
> from this sql file.
> 
> boss# gmake boss-install
> Checking DB schema...
> cd ./db && perl schemacheck
> Checking Site variables...
> cd ./db && perl sitevarscheck
> Use of uninitialized value in subroutine entry at
> /usr/local/lib/perl5/site_perl/5.8.8/mach/Mysql.pm line 137.
> Use of uninitialized value in subroutine entry at
> /usr/local/lib/perl5/site_perl/5.8.8/mach/Mysql.pm line 137.
> Checking Initial DB Fill ...
> cd ./db && perl dbfillcheck
> *** Your installed database is out of date with sql/database-fill.sql
>     To bring it up to date, run: 
>     mysqldump tbdb > tbdb.backup
>     (cd db; mysql tbdb <
> /usr/testbed/src/CRON_branch/sql/database-fill.sql)
> gmake: *** [install-dbfillcheck] Error 1
> 
> Any idea?
> 
> Best Regards,
> 
> Cheng Cui
> . Frey Computing Services Center . Baton Rouge, LA 70803
> 
> -----Original Message-----
> From: Cheng Cui [mailto:ccui1@tigers.lsu.edu] 
> Sent: Tuesday, March 23, 2010 12:57 PM
> To: 'Robert P Ricci'
> Cc: 'Mike Hibler'; 'testbed-admins@flux.utah.edu'; 'Leigh Stoller'
> Subject: RE: [Testbed-admins] Problem updating the database after git
> clone from emulab-stable
> 
> After I reconfigured the "config.status" to the right Python.h path,
the
> "boss> gmake clean & boss> gmake" has passed successfully. And I have
> finished mysql database update through step 6. Things look fine.
> 
> In step 7, I deleted all the temp tables. And re-synchronized
> "sitevariables" with our local settings in the "src/.. /sql/". Then, I
> deleted my "sitevariables" table and run the boss-install:
> boss> cd /your/objdir
> boss> sudo gmake boss-install
> 
> This time, another problem happens:
> 
> boss# mysql tbdb < /usr/testbed/src/CRON_branch/sql/database-fill.sql
> boss# cd ..
> boss# pwd
> /usr/testbed/obj/testbed-stable-obj
> boss# gmake boss-install
> Checking DB schema...
> cd ./db && perl schemacheck
> Checking Site variables...
> cd ./db && perl sitevarscheck
> Use of uninitialized value in subroutine entry at
> /usr/local/lib/perl5/site_perl/5.8.8/mach/Mysql.pm line 137.
> Use of uninitialized value in subroutine entry at
> /usr/local/lib/perl5/site_perl/5.8.8/mach/Mysql.pm line 137.
> Checking Initial DB Fill ...
> cd ./db && perl dbfillcheck
> *** Your installed database is out of date with sql/database-fill.sql
>     To bring it up to date, run: 
>     mysqldump tbdb > tbdb.backup
>     (cd db; mysql tbdb <
> /usr/testbed/src/CRON_branch/sql/database-fill.sql)
> gmake: *** [install-dbfillcheck] Error 1
> 
> I indeed updated tbdb by the database-fill.sql, but still get the same
> error. And what's more, I checked the table "sitevariables", it's
empty.
> But the "gmake boss-install" passed that check. Looks weird:
> 
> mysql> select * from sitevariables;
> Empty set (0.00 sec)
> 
> Any ideas? 
> 
> Best Regards,
> 
> Cheng Cui
> . Frey Computing Services Center . Baton Rouge, LA 70803
> 
> 
> -----Original Message-----
> From: Robert P Ricci [mailto:ricci@cs.utah.edu] 
> Sent: Tuesday, March 23, 2010 11:26 AM
> To: Cheng Cui
> Cc: 'Mike Hibler'; testbed-admins@flux.utah.edu; 'Leigh Stoller'
> Subject: Re: [Testbed-admins] Problem updating the database after git
> clone from emulab-stable
> 
> Thus spake Cheng Cui on Tue, Mar 23, 2010 at 11:05:20AM -0500:
> > No, there is no "config.status" file in the source code. Do you mean
> the
> > config.status in the object tree after configure?
> 
> Yes.