04.10.2019 15:50, Mark Rotteveel wrote:
> The problem is inherent to the monitoring tables. If you query one
> monitoring table, Firebird needs to collect and prepare a snapshot of
> the data for all other monitoring tables to be able to provide a stable
> and consistent view of the monitoring
04.10.2019 15:01, Marcelo Guimaraes wrote:
>
> Reading about FB performance, I hit this page:
> https://firebirdsql.org/file/documentation/reference_manuals/fblangref25-en/html/fblangref25-appx05-montables.html
>
> My surprise on the note was the warning: "In a highly loaded
> environment,
20.09.2019 13:16, Rustam rusta...@ukr.net wrote:
> I don't now API details, but my idea: what if we can specify 'mask' in
> Event name on clients, e.g. "ORDER_CHANGED*"?
>
> Then client subscribes to event "ORDER_CHANGED*", and his code receive
> all events, who name is starting with
19.09.2019 16:31, 'P-Soft - Fabio Codebue' wrote:
> Is there a way to fires event with some parameters?
Nope.
> Now I see that parameter-aware events should use a "pull"
> approach (parameter values are fetched by the client) instead of the
> current "push" one (data is unconditionally sent to
13.06.2019 13:43, 'Mr. John' mr_joh...@yahoo.com wrote:
> Using FB 3.0.4/Windows 10x64 ,I'm trying to delete SP source code as
> worked in 2.5:
> UPDATE RDB$PROCEDURES SET RDB$PROCEDURE_SOURCE = NULL WHERE
> RDB$SYSTEM_FLAG IS NULL OR RDB$SYSTEM_FLAG=0;
>
> bun now on FB 3 tried:
>
>
30.05.2019 9:35, liviuslivius wrote:
>
> i need to modify plan generated by Firebird and specify manual plan.
> I use Firebird 3.
> example on employee.fdb (sample from taken
> from README.common_table_expressions.txt)
>
> WITH RECURSIVE
>
13.05.2019 15:31, liviuslivius wrote:>> is TempCacheLimit is
really Integer or Int64?
int32 in v2.5, int64 in v3 and beyond.
Dmitry
07.03.2019 14:23, Carsten Schäfer wrote:
>
> With the left join FB does a full scan on Table_A and uses the Foreign
> Key index on Table_B. The index on table_B.field_B is not used at all.
>
> So you get millions of reads if the table contains millions of data
How many rows in table_B exist
28.02.2019 23:17, Issam Boughanmi wrote:
>
> i have managed to use the firebird embedded on the shared hosting by setting
> the FIREBIRD_LOCK variable
Why? This is not necessary.
> but i dont know if it safe to use this setup that way
It's safe as long as FIREBIRD_LOCK is defined as system
22.01.2019 20:54, Sean Leyne wrote:
>
> In my database I have some indexes with depth = 4 (db page size is
> already 16KB).
> The indexes are multi-segment, with the fields sequenced from least to
> most selective, and gstat reports a very low selectivity for the index
> (less than 0.0001).
01.06.2018 14:49, Mark Rotteveel wrote:
> Out of curiosity: why? Why not just allow any updates to NULL, even if
> the existing value already is NULL?
Setting NULL to NULL means the record wasn't changed. From the engine
POV it's the same as e.g. SET RDB$PROCEDURE_ID = RDB$PROCEDURE_ID, i.e.
01.06.2018 10:30, Chuck Belanger wrote:
>
> 2. The difference turned out to be that for one there was actually some
> source code in the fields. While the other that threw an error message,
> already had NULL fields in all records.
>
> 3. When limiting the UPDATE to a specific record rather than
16.05.2018 07:05, Javier Cintron wrote:
> Firebird ver 2.5.7.27050 64 bits
> Superserver
> Windows Server 2012R2
>
> What is the unit of TempCacheLimit?
Bytes. Settings > 4GB are not supported.
Dmitry
09.05.2018 20:02, liviuslivius wrote:
> Ok, i understand temp file caching but why Firebird create this temp
> file at all and how to prevent this?
Either some GTT was used or temporary blob storage was requested. Why do
you want it to be prevented?
> And second question, why file handle is
09.05.2018 17:15, liviuslivius wrote:
> >> No, caching is left up to OS.
>
> i do not understand above. I have dissabled os caching in firebird.conf.
> Also file is created by Firebird not OS. Can you explain this more?
You've disabled file-level caching for the database file, but this
09.05.2018 11:41, liviuslivius wrote:
> Is there particular settings to go always into RAM or some buffer
> settings which i can increase to avoid going to disc?
No, caching is left up to OS.
> I have now
> TempCacheLimit = 800M
It doesn't affect GTTs.
Dmitry
08.05.2018 20:50, Javier Cintron wrote:
> Firebird ver 2.5.7.27050 64 bits
> Superserver
> Windows Server 2012R2
>
> This link
> says that the maximum size of the page cache buffer is 131,072 pages.
>
> This limit also applies to Firebird 2.5 x64 superserver in Windows?
131,072 pages is the
09.05.2018 10:25, liviuslivius wrote:
>
> what is the purpose of temp file named e.g.
> fb_table_yfz4x5?
Storage for global temporary tables.
Dmitry
13.04.2018 13:30, acgomes2...@yahoo.com.br wrote:
>
> select N.DTEMISSAONOTA as DATA,
> 'DINHEIRO' as MEIO_PAGAMENTO1,
> cast('DINHEIRO' as char(10)) as MEIO_PAGAMENTO2,
> '00' as MEIO_PAGAMENTO,
> N.VLRTOTALNOTA as VALOR,
> 0 as AUTENT_01,
> 0
08.03.2018 12:23, jonatan.laurit...@yahoo.dk wrote:
>
> I have Firebird 2.1 dialect 1/charset NONE database and I am moving it
> to Firebird to get all the performance improvements (64 bit,
> multi-process capabilities, etc.). Move to dialect 3/charset UTF8 is
> quite involved, I am trying to
27.02.2018 06:42, hugo.lar...@yahoo.com [firebird-support] wrote:
>
> Does null values take up disk space?
Yes, but much less than the declared column size.
> If I have a table with 4 VARCHAR columns and I add 4 more does the table
> need double space even if the new columns are null?
Already
21.02.2018 15:46, Alexander Tabakov wrote:
>
> What happens in case 3 (stored in the same database) when we move the
> database on a different server. Shall credentials stored in place take
> precedence over the other server's configuration?
It depends on the databases.conf on the other
16.02.2018 22:23, Helen Borrie wrote:
>
> A gbak backup does a sweep by default, unless you run it with the -g
> switch ( [no_]g[arbage-collection] ).
Small correction: gbak performs a global cooperative GC pass, not the
sweep (which main purpose is moving OIT).
Dmitry
13.02.2018 05:07, Hamish Moffatt wrote:
> I attached gdb and dumped the state of all threads, which shows every
> one of them is waiting on a mutex or futex, ie it's dead locked. I've
> attached the trace.
You need to download the debug symbols (Firebird-debuginfo-* package)
and copy them into
10.02.2018 22:33, Aldo Caruso wrote:
>
> A strange behavior is seen in the combination not null and no default
> value. It is returned as a 0 for selects but treated as a NULL when
> comparing.
In fact, the engine returns NULL. But query prepare describes the output
descriptor as NOT NULL. Some
03.02.2018 23:14, marco andreolli wrote:
>
> but RDB$SET_CONTEXT() it's not right for me, because it store
> information in actual connection (or transaction) and they can not be
> read out of this (with a monitor program for example).
It can be, via MON$CONTEXT_VARIABLES.
Dmitry
08.12.2017 13:56, Mark Rotteveel wrote:
>> Because you are sending queries in connection charset. And string
>> literals in them are
>> in connection charset. Because of that your 'x' cannot have COLLATE
>> PXW_HUNDC.
>
> I don't agree with this. The string literal is in the connection
>
24.11.2017 10:53, alauren...@gmail.com wrote:
>
> I have a query with no results and I doo't understand what I do wrong:
>
> SELECT DENSUB, CODSUB from SUBANS,TEMPSUB
>
> WHERE SUBANS.CODSUB=TEMPSUB.CODSBB AND SUBANS.CODSUB NOT IN (SELECT
> CODSBB FROM TEMPSUB) AND SUBANS.CODPF='FL1' order by
29.09.2017 09:50, Ertan Küçükoğlu wrote:
>
> Answer to your question is yes. KAYNAGAGORE and SEBEBEGORE are not
> indexed. Please note that they're not indexed on PostgreSQL, too.
>
> Question remains as how PostgreSQL can cope with identical data only
> with primary key indexes faster than
29.09.2017 01:08, Ertan Küçükoğlu wrote:
Answers can be found in the plans:
> PLAN JOIN (RAPOR_EK ORDER rapor_ek_pkey, SORT (DT M NATURAL))
This join plan is terrible for FB. Are KAYNAGAGORE and SEBEBEGORE
indexed? They seem not.
> Merge Left Join (cost=6600.19..6610.98 rows=560 width=340)
10.09.2017 02:13, Daniel Miller wrote:
> The misunderstanding is probably mine - but I've never heard of 34-bit
> values/precision before. Is that a typo for either 32 or 64? Or "34"
> is correct?
34 *digit* precision. If "34-bit" is mentioned there, then it's a mistake.
Dmitry
03.08.2017 20:33, Jesus Garcia wrote:
>
> I only see three items pending, but anyway, why not remove Nbackup from
> firebird until it is safe to use?
>
> Why is there a tool that cannot be used in a safe way?
AFAIK, there are no known critical issues with nbackup in latest FB
releases
28.03.2017 10:06, Martijn Tonies wrote:
>
> So the first 'fetch' operation on a position that's invalid returns
> "rowcount=0",
> any additional fetch (which includes a move) returns an exception?
Yes. Moving cursor from "last" to EOF is perfectly valid, "no data"
condition is not an error (as
27.03.2017 11:23, 'Martijn Tonies wrote:
>
> The FOR SELECT loop returns 1..10 as the values.
>
> The fetch/suspend lines return 1..10, with a row count of 1, and another
> ‘10’ with a row count of 0.
>
> If I add another FETCH, I get this error while executing the routine:
>>>attempt to fetch
14.03.2017 20:54, 'Leyne, Sean' wrote:
>
> The -o -w switches generated details such as:
>
> OWNER BLOCK 732776
> Owner id: 114332029419524, type: 1, pending: 0
> Process id: 26620 (Alive), thread id: 2348
> Flags: 0x08 wake
> Requests (431): forward: 732888,
12.03.2017 00:12, 'Leyne, Sean' wrote:
>
> Am I right to think that I need to create a process to run the command on a
> regular basis (every 5 secs?) to find what objects locks are waiting for?
Yes. But we're still guessing in the dark, the reason of the slowdown
could be completely unrelated
11.03.2017 23:56, Dmitry Yemanov wrote:
>
> I asked for -w, not -iw.
Or maybe -o -w, if just -w prints nothing.
Dmitry
++
Visit
11.03.2017 22:40, 'Leyne, Sean' wrote:
>
> Here are some initial results
>
> C:\FIREBIRD\bin>fb_lock_print.exe -d x -iw 1 5000
I asked for -w, not -iw.
Dmitry
11.03.2017 21:03, 'Leyne, Sean' wrote:
>
>>> We have a client with 320GB database (running FB CS v2.5)
>>
>> Is it really so? FB does not support LockHashSlots more than 64K, it would
>> truncate your 90001 down to 65521.
>
> Is the engine that smart to trim to an exact prime number?
Nope, 65521
10.03.2017 23:58, 'Leyne, Sean' wrote:
>
> We have a client with 320GB database (running FB CS v2.5)
Is it really so? FB does not support LockHashSlots more than 64K, it
would truncate your 90001 down to 65521.
> Yesterday, I ran fb_lock_print to check on the database and found a
> “Mutex wait”
06.03.2017 21:48, 'Neil Pickles' wrote:
> One thing I don’t understand is why there would be over 744 million txns
> on this file? Is a new txn started each time a connection to the
> database is made?
Of course, the server needs to validate your login/password against the
security database.
>
01.03.2017 11:37, jonatan.laurit...@yahoo.dk wrote:
> Actually, I don't understand this thing about old and new authorization
> system. Does it mean that there are two SYSDBA and other user accounts -
> one for old and other for new authorization system and does it mean that
> there are two sets
01.03.2017 11:58, jonatan.laurit...@yahoo.dk wrote:
> OK, I have working Firebird 3.0.1 installation and I have copied
> security3.fdb from the broken Firebird installation and now I am also
> receiving the error message. So - security3.fdb is somehow broken,
> problem is in security3.fdb
It's
21.02.2017 15:53, michael.vilhelm...@microcom.dk wrote:
>
> ANS4005E Error processing
> '\\db07\c$\ProgramData\firebird\fb_lock_068acd221200a400': file
> not found
>
> What does this mean?
>
> Should I be worried?
Nothing to worry about. Some MS or 3rd party software (backup?
antivirus?)
16.02.2017 21:20, liviuslivius wrote:
>
> can this plan be selected by Firebird in somehow automatically without
> going to left join or +0 solution?
In this particular case, I'm afraid you don't have any other choice. But
I will think a little bit more, maybe some clever idea will come.
17.02.2017 15:29, André Litfin a.lit...@xsigns.de wrote:
>
> I thought FB3 should remove the limitation to max. length of for
> varchars (32765 bytes). I tried today to create a longer varchar field
> without luck. Is the limitation still active?
Yes, there were no plans to remove it.
Dmitry
16.02.2017 21:20, liviuslivius wrote:
>
> Firebird3
There's no such a version.
> plan is "expected"
> PLAN JOIN (SD ORDER IXDU_SENSOR_DATA__ID, S INDEX (PK_SENSOR))
What's definition of IXDU_SENSOR_DATA__ID?
Dmitry
09.02.2017 17:39, hamacker wrote:
> Well, the man page says fully supports text BLOBs, I assume that binary not.
For blobs, it works with raw binary data.
Dmitry
09.02.2017 14:48, hamacker wrote:
>
> There is a HASH function but only for string.
It works for blobs as well.
Dmitry
06.02.2017 22:08, Dimitry Sibiryakov wrote:
>
> When a string is sent to server via a query parameter, how is its character
> set
> determined: from XSQLVAR.sqlsubtype field or from connection's lc_ctype
> parameter?
From the XSQLVAR.sqlsubtype, which in turn is determined based on the
23.01.2017 15:05, Kjell Rilbe wrote:
>
> I'm still on FB 2.5.2, so... has this been improved in 3.0?
Yes, it was (to some extent).
Dmitry
29.12.2016 23:05, Helen Borrie hele...@iinet.net.au wrote:
>
>> On windows systems, if you create a database without path specification,
>> firebird tries to create it in the windows/system32 directory.
>
> No; it will use the current directory
By default, for any Windows service (including
22.12.2016 10:43, Geoff Worboys wrote:
>
> I do have a couple of EXECUTE STATEMENT ON EXTERNAL calls in
> the database code. I hadn't given them a thought. In theory
> none of them should happen during the startup, but it does
> seem the most likely explanation.
They don't need to run during
21.12.2016 03:45, Geoff Worboys wrote:
>
> Firebird v2.5.6 32bit "Classic"
> on Windows 64bit 2012 R2 Standard
>
> I have an ON CONNECT database trigger in which I try to collect
> the client address and network protocol using:
>
> rdb$get_context('SYSTEM', 'CLIENT_ADDRESS')
> and
21.12.2016 12:39, Roland Turcan wrote:
>
> But still, why this problem occurred? Why didn't some generators have
> it =1?
Could that database be migrated from some pre-release (e.g. Beta) v3
build without backup/restore?
Dmitry
06.12.2016 11:50, Mark Rotteveel wrote:
> I noticed something interesting, when restoring a Firebird 2.5 database
> to Firebird 3.0, only user generators (RDB$SYSTEM_FLAG = 0) get a value
> of 1 in RDB$GENERATOR_INCREMENT, system generators get 0. Is that
> intentional?
System generators are not
06.12.2016 11:05, Roland Turcan wrote:
>
> I have installed new clean database to new customer having FB3.0.1
> 32bit running on Windows7Pro 32bit. He was complaining that he is not
> able to add new record into some parts of our software. I have
> connected to his computer and
29.11.2016 13:45, 'Israel Pinheiro' wrote:
>
> Invalid Token. Dynamic SQL Error. Invalid Command Data Type unknown
Check datatypes of all union parts, i.e.
select A1, B1, C1
union
select A2, B2, C2
union
select A3, B3, C3
Data types for (A1, A2, A3) should match exactly. The same for B and C.
04.11.2016 13:30, Si Carter wrote:
> The memory usage crept up to just under 3gb again
Once again, where do you get that number from? Task manager? Perfmon?
top? Screenshot proof would be appreciated.
> The memory usage crept up to just under 3gb again, and I took a snapshot
> of the data
04.11.2016 13:48, Norbert Saint Georges wrote:
>
> the recording limit for FB3 is the same as FB2.5?
Yes.
Dmitry
++
Visit http://www.firebirdsql.org
03.11.2016 09:37, Si Carter wrote:
>
> I recently installed FB v3.0.1.32609 on Windows server 2008.
>
> This is a dual installation as I still have 2.5.3 installed.
>
> I noticed that my fb3 service is using in excess of 4gb memory
How did you get that number?
> Given this sql
>
> SELECT
24.10.2016 18:08, 'Gregor Kobler' wrote:
>
> Is it also valid for Server 3.0? Are the roules 4 transaction per byte
> also valid at FB3 ?
Yes.
Dmitry
16.09.2016 09:02, 'Gregor Kobler'wrote:
>
> When you querying the MON$STATEMENTS table, the active statements are
> not shows a transaction-id and not a timestamp, These columns are allwas
> Zero / NULL!
Transaction ID and timestamp are reported only for active statements,
they're NULL for idle
06.09.2016 08:55, Tony Christiansen wrote:
>
> My client needed a server rebuild after a ransomware attack so I took
> the opportunity to upgrade from FB2.5 to FB3. My client application is
> Delphi using IBX components.
>
> Now they are experiencing repeated crashes with an array of exceptions
>
13.07.2016 12:01, Norbert Saint Georges wrote:
> it's normal ?
Yes, scrollable navigation is not yet supported by the remote protocol,
only by the embedded engine. Maybe support will be added in some point
release.
Dmitry
16.06.2016 09:02, Roland Turcan wrote:
>
> I am trying to deploy Firebird 3.0 to our customers. Only three small
> customers are running on FB3, but all of them have the same problem:
>
> invalid transaction handle (expecting explicit transaction start).
>
> This exception happens on different
07.07.2016 10:54, jonatan.laurit...@yahoo.dk wrote:
> Are the following statements correct?
>
> - Most Firebird client software still is 32bit (Flamerobin, IBExpert,
> Delphi 2009 and older editions) and therefore it requires 32bit Firebird
> 3.0 client dlls (fbclient.dll or gds32.dll) even if it
07.07.2016 07:25, Alexandre Benson Smith wrote:
>
> I am looking at the per-database configuration (FB 3) and have two doubts:
>
> I defined my databases.conf like:
> demo_naolocal=c:\bd\demo_naolocal.fdb
> {
> Providers = Remote,Loopback
> }
>
> after that on the same machine I tryed:
>
>
23.06.2016 19:39, 'Israel Pinheiro' wrote:
>
> I created a view that works perfectly on FB 2.5, but now need to create the
> same view on FB 1.5, but appears error in creation. Can help me?
http://www.firebirdfaq.org/faq92/
Double-check that both union parts have exactly the same datatypes for
22.06.2016 13:44, Martijn Tonies wrote:
>
>>> When connected to the database from another application, here:
>>> C:\Program Files\Firebird\Firebird_3_0>gbak -b -m -t -v
>>> -user sysdba -pas masterkey
>>> "e:\temp\2014 FB Conference.fdb" "e:\temp\2014 FB Conference.bak"
>>
>>Don't you forget
22.06.2016 13:40, Dmitry Yemanov wrote:
> Worth registering in the tracker.
It seems it's already there:
http://tracker.firebirdsql.org/browse/CORE-5263
Dmitry
22.06.2016 13:14, Martijn Tonies wrote:
>
> When connected to the database from another application, here:
> C:\Program Files\Firebird\Firebird_3_0>gbak -b -m -t -v
>-user sysdba -pas masterkey
>"e:\temp\2014 FB Conference.fdb" "e:\temp\2014 FB Conference.bak"
Hostless connection string,
22.06.2016 12:49, 'Martijn Tonies wrote:
>
>> I understood you as that regular gbak does not work when the database is
>> in use, so now you try the -se switch.
>
> That is correct, but as you can see, using -se resulted in an error.
So what was the error using gbak without -se? And what was the
22.06.2016 12:35, 'Martijn Tonies wrote:
>
> As per original example:
> Here’s the command:
> C:\Program Files\Firebird\Firebird_3_0>gbak -b -m -t -se service_mgr –v
> -user sysdba -pas masterkey
>"MARTIJN-WS-DEV/3054:e:\temp\2014 FB Conference.fdb"
>"e:\temp\2014 FB Conference.bak"
>
22.06.2016 11:25, 'Martijn Tonies wrote:
>
> Firebird 3 fails to back up when the database is in use
It does not, as long as you don't use a "hostless" connection string.
Dmitry
21.06.2016 03:52, Seyeong Jeong wrote:
>
> I have a firebird database with ODS version 10.1. While I can open this
> with server version WI-V2.1.7.18553 Firebird 2.1 on Windows without any
> errors, the same operation fails on Ubuntu.
I suspect your Windows version (or Firebird version running
17.06.2016 10:56, Dimitry Sibiryakov s...@ibphoenix.com wrote:
>
>>> Is there the limitation of 31 bytes for column names in the new FB 3?
>>
>> 31 CHARACTERS - as before
>
> No, it is 31 byte.
Correct, the limit is 31 *bytes*. The legacy UNICODE_FSS charset is
multi-byte (up to three bytes per
15.06.2016 22:48, 'livius' wrote:
>
> we can write sql like
> select * from table_name but really this is
> select * from schema.table_name
> and if i write this in stored procedure – i suppose that “object_id”
> will be stored in blr(when schemas will be avaiable)
BLR will contain
15.06.2016 22:49, 'livius' wrote:
>
> but this feature is considered for FB4
No, it's not. At least not for v4.
Dmitry
13.06.2016 00:18, 'livius' wrote:
>
> what are + and – with working with schema?
> What benefits are between
> schema_name__table_name and real schema implementation?
> schema_name.table_name
Ability to have a default schema (per user, per connection). So that you
may have multiple completely
14.06.2016 15:52, fabia...@itbizolutions.com.au wrote:
> 1) The variable "RemoteBindAddress" in firebird.conf is a Server Wide
> value, or it can be set per database in databases.conf?
Server wide.
Dmitry
08.06.2016 11:36, m.djo...@yahoo.com wrote:
>
> First I thought it's because of the results' type after the different
> operations, but even in cases when the result before ROUND is a DOUBLE
> PRECISION (according IBExpert) the round gives different results if the
> NULL in the COALESCE function
03.06.2016 17:59, Svein Erling Tysvær wrote:
>
> I simplified the query to
>
> WITH TMP( TJENESTEENHETRESHID, TJENESTEENHETLOKAL, IMPORT_DATO ) AS
> ( SELECT TJENESTEENHETRESHID, TJENESTEENHETLOKAL, MAX( IMPORT_DATO )
> FROM NPR
> WHERE TJENESTEENHETRESHID > 0
> AND
03.06.2016 11:27, Svein Erling Tysvær wrote:
> This query:
> WITH TMP( TJENESTEENHETRESHID, TJENESTEENHETLOKAL, IMPORT_DATO ) AS
> ( SELECT TJENESTEENHETRESHID, TJENESTEENHETLOKAL, MAX( IMPORT_DATO )
> FROM NPR
> WHERE TJENESTEENHETRESHID > 0
> AND TJENESTEENHETLOKAL > ''
> GROUP
03.06.2016 05:24, trsk...@yahoo.com wrote:
>
> 1) Since FB 3.0 support WireCompression, should I still use Zebedee?
No point anymore.
> 2) Is that possible to hide stored procedure/trigger codes in FB 3.0 ?
> Usually I used this statement :
>
> UPDATE RDB$TRIGGERS a set a.RDB$TRIGGER_SOURCE =
02.06.2016 07:42, trsk...@yahoo.com wrote:
>
> I have a table with existing data & got a problem to add field that must
> have a value -> alter table Tbl1 add Fld1 int not null.
What value must it have? Add a default clause to this statement. Or add
the field as nullable, populate with data and
02.06.2016 03:44, trsk...@yahoo.com wrote:
>
> Am I correct that maximum allowed RAM for database cache on SS FB 3.0 in
> Win64 is only 2 GB? Calculated from max DefaultDbCachePages (131072) x
> max Page Size (16384) = 2 GB.
Correct for 32-bit builds. 64-bit builds don't have such a limit.
>
01.06.2016 04:16, Fabian Ernesto Chocron wrote:
>
> We are having trouble setting up the database cache size on FB 3.0
> running on Windows 2008 R2 64 bits with 32 GB ram.
>
> The problem we have is we cannot get the server to allocate the ram for
> the cache as we intend. With FB 2.54 we had the
26.05.2016 14:50, dondondonje...@yahoo.com wrote:
>
> "C:\Program Files\Firebird\Firebird_3_0\gbak.exe" -b -v -t -se
> service_mgr employee e:\xxx.bak -user SYSDBA
> works if no one is connected else i always get error "file is used by
> another process".
gbak.exe -b -v -t -se
23.05.2016 17:22, 'Heron Vieira' wrote:
>
> I'd like to translate firebird.msg, but I think I hit a wall with the
> size of the messages.
>
> So far I translated some messages by editing the file in a HEX editor,
> but I think the correct way would be modifying the source code, but I've
> no idea
23.05.2016 22:32, Alexey Kovyazin wrote:
>
> 44. This option turn off individual savepoints.
Individual savepoints are never turned off. This option disables the
root (transaction-level) savepoint used to transform a rollback into a
commit (by undoing all the transaction changes).
Dmitry
01.05.2016 19:49, trsk...@yahoo.com wrote:
>
> This is the Query plan from SS FB 3.0
>
> PLAN (GET_STCK NATURAL)
This is just a different output for select from procedure GET_STCK, v3.0
does not show all its internals as v2.5 did.
You'd better compare per table performance counters
29.04.2016 17:57, Dmitry Yemanov wrote:
>
> Did you see the same issue on v2.5 or only on v3.0? Maybe it's a
> regression introduced in the new version. I'm checking the code now,
> will report back if anything useful is found.
I see a possible reason in the code, it needs testing. Can you send
29.04.2016 12:38, Gabor Boros wrote:
>> This is true for two concurrent transactions but may fail for three or
>> more concurrent transactions.
>
> Your words shocked me. Firebird cannot handle more than two concurrent
> transactions? What is the purpose of WITH LOCK if see deadlock with and
>
28.04.2016 16:19, Gabor Boros wrote:
> In the language reference for "isc_tpb_read_committed + isc_tpb_wait" I
> see this: "Update conflict exceptions can never be raised by an explicit
> lock statement in this TPB mode."
This is true for two concurrent transactions but may fail for three or
25.04.2016 10:40, Behlül Şimşek wrote:
>
> is it possible to protect our database with a user/password in FB 3.0.
> if db move out the another pc, won't open without knowing correct
> user/password.
You may protect it with encryption. Usre/password cannot protect a
publicly distributed database,
14.03.2016 10:41, liviuslivius wrote:
> Why?
> Index is a Tree? And if i found VALUE 2 in A key then i can fast find
> value 5 in sub key (leaf)
> You scan throught keys in A, and then in finded nodes you look for leafs
> in B
Compound index key is a single concatenated value, not two separate
14.03.2016 10:32, liviuslivius wrote:
>
> simple table
> CREATE TABLE XXX(
> A INTEGER
> , B INTEGER
> , CONSTRAINT UK_XXX__A__B UNIQUE(A, B)
> )
> commit;
> SELECT * FROM dbo.XXX X WHERE X.A BETWEEN 2 AND 30 *AND* X.B BETWEEN 5
> AND 60
> Select Expression
> -> Filter
> -> Table
11.03.2016 13:39, liviuslivius wrote:
>
> I see that in FB3 this was changed and before apply not null to column -
> all existed data are validated.
Correct.
> Is three some possibility to disable this behavior?
Nope.
> On my big table with 910 468 710 records this operation take 72 minutes.
>
09.03.2016 05:11, russ...@belding.co.nz [firebird-support] wrote:
>
> Adding UDFs to databases in FB 3.0 RC2 is giving an error when using any
> function in the added UDF. The UDFs packaged with FB3 are OK.
>
> To the EMPLOYEE db packaged with FB3.0 RC2 I can add UDF fbudf.dll,
> which is packaged
1 - 100 of 189 matches
Mail list logo
