logs archiveIRC Archive / Freenode / #oracle / 2009 / November / 23 / 1
PocketKnives
How would some people in here create an inventory database that needs to have multiple user levels, and a front end "user-friendly" application with a login
Techdeck
hey guys, anyone knows what the HELL does this mean? ORA-01654: unable to extend index TMUSIC_CORE.DR$T_ALBUM_NAME_IDX$X by 8 in tablespace TMUSIC_CORE_TAB_TBS
dgs_
i'm trying to run some reports under linux - i gather i need to use rwrun.sh or rwclient.sh, but they don't seem to have been installed with the database (11g). anyone know what i need to install to get these?
Dormir
reports is part of developer
along wirh forms
dgs_
k.
ta. i think i've found the installer - trying it now
if only this stuff was in the oracle debian repository =)
Yuyp
hey... how can i write a query that would generate a one-column dataset with incrementing integer values? like (0), (1), (2), ..., (n)
is this possible?
Maxel
can anyone help me with pl/sql, specifically using a nested block?
placydo
Hey guyz! My oracle account were blocked because of too many incorect passwords
how to unblock it (im tottaly lame)
TheBonsai
ALTER USER <user> ACCOUNT UNLOCK;
         

placydo
many thanks
FrEaKmAn_
hi, Im looking for tools to sync databases
I only found tools for syncing calendar
I just wonder if its possible to sync parts of database to another database
I see that Oracle Streams should solve this
placydo
TheBonsai when i do that it goes to next line and writes
2
xenet
placydo end the command with ;
placydo
cheers
bah i cant login into db
tu unlock account
to*
xenet
can you login locally with sqlplus?
by using the command: sqlplus "/ as sysdba"
placydo
thanks
im done
luv yo ;)
xenet
you're welcome
mastro
is there a documentation on the weblogic deployment process? I've an issue deploying a third party application: it does deploy just fine if I start the server the first time, but it give errors if I update the webapp instead. It's some weird issue with class loaders ... I need to understand what weblogic exactly do on those cases to study the issue. Thanks
actually.. what I need is to understand the classloader hierarchy of a weblogic deployment, when and how different library got loaded and so on... what's different in that if I start the server or I just upload the war and so on
xenet
mastro: maybe you can find some info on http://download.oracle.com/docs/cd/E13222_01/wls/docs100/deployment/index.html
mastro
xenet: thanks. I was already reading this: http://download.oracle.com/docs/cd/E12840_01/wls/docs103/programming/classloading.html
xenet
yes, that seems to be exactly what you where looking for.
mvno_subscriber
hi, i'm having problems installing the oracle odbc in windows. I already have the instant client which i use for Toad, but when trying to reverse engineer a schema using visio i encountered problems getting the correct odbc driver. I downloaded the driver from Oracle (a zip file with some drivers and instructions on where to place them), but that doesn't really help... it doesn't show up under my odbc sources. is there a simple step-by-step guid
usn_work
good CET (after)noon
cmug
shared servers can be enabled dynamically right? Does it kill current DEDICATED connections or are they handled until they are finished?
usn_work
cmug, the choice between dedicated- and shared serv3ers is done by the client, for example in the tnsnames.ora
and you can't disable dedicated connections
cmug
yeah, but my client config is set to use shared
and my server is set to 0 shared_servers, but changing that does not require a bounce?
usn_work
nots, it will start up the shared servers within one minute
s/nots/no
cmug
ok, i'll change it then
usn_work
don't forget to configure a dispatcher properly
         

cmug
It probably won't touch the current sessions?
the suggested amount of dispatchers was 10 sessions per dispatcher if I remember correctly
Number of dispatchers = CEIL ( max. concurrent sessions / connections for each dispatcher )
no it was the number for shared_servers that was 10sessions/server
tiny
Hello! I'm getting weird insert error:
insert error: ora-01578: ORACLE datablock corrupted file # 1024 block # 59108936
file number 1024?
hali
panic!
you need to recover the datafile
tiny
hali?
hali
from a backup and replay the archive log
file number 1024 looks odd though
unless you actually have loads of datafiles
tiny
hali: yes, funny thing is other "replicator" processes are using the same file, same tables and all is good
hali: nope just one DB file
hali
if it's production system i'd log a call with oracle
at least if you can't have that node go down
tiny
ok
hali
you could take a copy of the file and run dbv on it
is the storage under the file ok? raid etc
tiny
file is huge 600GB, I need another USB HD unit to make copy
storage is ok
HW RAID5
hali
damn
that will be a pain to recover
tiny
It's been ages since I've used metalink ... it's asking me for "Support Identifier". Any ideas what that is?
hali: tell me about it :)
hali
i usually cap datafiles at 32gb unless it's a huge db
tiny
hali: I'll know for the future
hali: can I use dbv to only focus on particular blocks of file?
hali
can't remember
tiny
just did
it found something broken but it didn't fixed it ... I'm assuming I can run this on a "live" database?
hali
i think it should be a file copy or shutdown
i haven't used dbv in ages
tiny
I see ... well it says: Page ##### is influx - most likely media corrupt
fsck should report this
but it didn't
ok .. .this means I need to shut it down
hali
how do you do backups?
file copy of a 600gb file will take time!
tiny
hali: There's no backup on this one yet. This is a central DB that serves as a replicator point from remote DBs which are backed up with rman.
Why no backup? Customer didn't provide media.
hali
handy
tiny
yep
hali
you need like $50k for storage
tiny
why so mcuh?
err much*
hali
you probably need 20k for a disk array with enough space for a couple of copies of that file
and say 15k or so for a tape library and software
tiny
oh that ... well it would be nice to have that yeah
tape lib we have
only client for this one ... but it sucks
ARcserve
hali
ugh, i hate arcserve, but i haven't used it for like 7 years, probably better these days
im 100% vertias netbackup
tiny
thanks for the tip, I'll recommend when they order SAN storage next year
hali: uhm, so if there's just one block that's corrupted and it's identified as a blob (which lets say is expendable in our case) can I just delete this block somehow from our DB?
hali
don't think so
tiny
What I don't get is other 7 replicator processes runing just fine and are inserting into the same datafile/tables/tablespace
Dormir
tiny: was the database restored from a backup?
tiny
No Dormir
Dormir
ok, thought you might've had a nologging corruption.
tiny
i see
Dormir
which isn't really corruption, but gets reported that way
tiny
there's no way to "bypass" or mark a certain block as corrupted
so DB doesn't use it anymore?
hali
the easiest way would be to use rman to do a blockrecover, if you had rman backups :-)
tiny
hali: that will probably be my final answer to a client :)
aae
No body can help me resolving this error "Checking monitor: must be configured to display at least 256 colors
>>> Could not execute auto check for display colors using command /usr/X11R6/bin/xdpyinfo. Check if the DISPLAY variable is set. Failed <<<<
"
hali
aae: yes, sort it out
oracle needs an X11 display
aae
hali: can u show me how to fix it??
hali
aae: are you installing on a local machine?
aae
yes
hali
ssh -X oracle@localhost
is the easiest way to get around it, without learning xauth
ssh as the user who owns X that is
aae
so connecet via ssh to my machin then install oracle from ssh ?
hali
just to get $DISPLAY and xauth setup, yes
aae
hali: can you explain to me the error where come from?
Ok!!
Egyptian[web]
afternoon all
i m at a loss .. i m on my upteenth attempt at installing a 10gR2 rac and i keep getting this message Expecting the CRS daemons to be up within 600 seconds. Failure at final check of Oracle CRS stack. 10
the documentation i found tells me to duh .. put a node name instead of null .. but its the /home/crs/root.sh that brings up the error
help ?
usn_work
first: Why 10g, not 11g?
Second: Did you make sure your SSH passwordless connectivity to all remote AND the local node?
Egyptian[Home], ping
Egyptian[web]
usn_work: project requirements
hi usn_work
usn_work
For your project manager: Basic Support for 10g ends in 2010
But what about ssh?
Egyptian[web]
ssh user equivalence is set up
node 1 refuses to start while node2 starts up perfectly
usn_work
did you start root.sh on node 1 first, and did you wait until it finished?
Egyptian[web]
hope you will forgive me but if i disappear at times (baby in the room)
usn_work: on node1 first yes.. error is from node1 .. no error on node2
usn_work
Can you copy/paste the error message please?
if I disappear from time to time: Work in the room *g*
Egyptian[web]
sec
http://fpaste.org/s6S0/
tiny
I want to report problem with metalink but it stops at error "Select a category" which I can't select since it's shaded.
So I can't report a problem.
usn_work
did you use the flash version?
tiny
yes, new one
usn_work
Try without flash: http://www.usn-it.de/index.php/2009/11/10/new-my-oracle-support-ex-metalink-use-without-flash-html-option/
Egyptian[Home], you have re-run your root.sh. This is not possible
OCR has to be blanked (with DD or similar) to re-run the script
stop CRS on all nodes and blank OCR in the whole
after that, rerun root.sh on node 1 and WAIT UNTIL FINISHED. If it fails, paste the error message, DO NOT continue to node 2
tiny
usn_work: TY! I'll try that one.
l3thal_
usn_work: people still use forms 6i FYI :)
usn_work
l3thal_, yes, but for new projects as well? ;)
I doubt it
l3thal_
i dont
i see it with my own eyes
people use it
usn_work
*shiver*
l3thal_
usn_work: even universities , the ones that really dont care about $$$ :D
usn_work
Egyptian[Home], ping
Egyptian[web]
usn_work: thanks ... i noticed one of my mistakes was running root.sh on both nodes .. but dint know how to recover
usn_work
just blank :)
Yuyp
hey... how can i write a query that would generate a one-column dataset with incrementing integer values? like (0), (1), (2), ..., (n)
is this possible?
l3thal_
will there be oracle forms 11 any soon ?
tiny
Following resolution paths on oracle support pages I determined that I have a corrupted index. But when I try to display or identify that index I get nothing ie no rows.
SELECT tablespace_name, segment_type, owner, segment_name
FROM dba_extents
WHERE file_id = &AFN
and &BL between block_id AND block_id + blocks - 1
;
That gave me: TS_NAME: CARINATS, SEGMENT_TYPE: INDEX, ...
so it's index right?>
I've inserted my AFN and BL numbers
usn_work
this SQL has nothing to do with corruption, it's just to identify an object by block ID
tiny
usn_work: yes and it's an index
usn_work
if you get no lines, you selected a block Id that's not there
tiny
I get no lines when I execute
SELECT owner, constraint_name, constraint_type, table_name
FROM dba_constraints
WHERE owner='&TABLE_OWNER'
AND constraint_name='&INDEX_NAME'
;
to determine constraint type
usn_work
so there is no constraint enforced by this index
tiny
usn_work: hmm .... I guess I got sidetracked with the text that followed:
Possible values for CONSTRAINT_TYPE are:
PThe index supports a primary key constraint.
UThe index supports a unique constraint.
« prev 1 2 next »