Both "dump database" and "sqlmaint.exe -bkupdb" would be able to backup database, one important advantage I can see from "sqlmaint" is that it has the option to cleanup the very old generations of backups by using -delbkups option in a scheduled maintenance backup job, while the "dump database" would first create a dump device and it'll grow with time.
Is there any suggested way to use the "dump transaction" but still we can get control on the size of dasd used ?
Any negative part of using "sqlmaint" utitity to backup database as compared to the "dump database " utility ?
Anthony
We run SQLMaint hourly ( as a scheduled task ) to clear our transaction logs. Last night our hard drive filled up during a log dump, causing SQLMaint to fail, and a subsequent transaction filled up the log.
We succeeded in manually truncating and clearing the log, but SQLMaint still insists on us doing a DUMP DATABASE to clear the logs before it will run again. Unfortunately, we do not have enough space left on our hard drive to do a full database dump.
Is there any way of fooling SQLMaint into believing that a DUMP has been run? Some flag in a system table somewhere, or maybe a registry entry?
1. I need to create a replication enviroment. In that i want to snapshot the database as and when i want from my Java program, so that selective tables are updated with latest data. Is there anyway i can do with out creating native replication methods.
2. Is there any way with out creating a snapshot or replication. I need to copy the data from my master database to child database(my reporting database) with out straining much my master database
When I use "DUMP DATABASE", it will help me to backup both database and transaction log. However, it will not help me to clean up the transaction log. Therefore, I am worry that will the following happen?
1. DUMP DATABASE Start backup database ( there are transactions during the backup process ) Finish backup database Start backup transaction Finish backup transaction
2. DUMP TRANSACTION Start backup transaction and clean up the transaction log
So, will the transactions been backuped twice? If yes, will it cause those transaction do twice when I restore the database then restore the transaction?
Tried googling, but I guess I have the wrong terminology for MSproducts.Using PostgreSQL, or MySQL, if I wanted to tranfer a database from onemachine to another, I'd do a "dump" with a single command, such as:mysqldump [databasename] > [path/filename]Then I'd take the file generated and dump it back onto a new machine.What's the process with MS SQL Server? Can someone point me in theright direction to a webpage dealing with this?--[ Sugapablo ][ http://www.sugapablo.com <--music ][ http://www.sugapablo.net <--personal ][ Join Bytes! <--jabber IM ]
I'm a newbie. I've installed a replication between 2 NT4 servers, with SQL 6.5. My logreader tasks has crashed "Can't allocate space for objects 'syslogs' in database 'distribution' etc...". With a DBCC SQLPERF(LOGSPACE) I can see that the log is 100% full, but I can't dump it : when I use DUMP TRAN DISTRIBUTION WITH NO_LOG, my distribution log is still 100% full. So how can I do to dump it and what can I do to prevent it from filling up ? Do I need to dump it like I do for my database but how could it work if a DUMP TRAN doesn't work ?
Thanks in advance.
Sorry for the English, it's not my native language.
I am fairly new to SQL server and am having a problem getting the SQL backup or "dump" to do what I need it to do. My desired end result is to have a database backup every night to a backup device on the same server as SQL is running, and keep one weeks worth of databases in the backup device, overwriting the database backup from 7 days before. This device is then transferred sometime later to another server using the NT AT scheduler. The first time I tried to accomplish this I used Enterprise manager to schedule a full backup to be run each night at midnight. The backup device was set to be initialized and retained for 7 days. This procedure worked once, the first night. There after it failed. In looking at books on line it said that I needed to change the transact -SQL command from INIT to NOINIT as the INIT instruction was the cause of the failure. I changed the statement in the Scheduled Task manager to NOINIT. That seemed to do the trick as I checked the backup device each of the next few days and saw that it was indeed performing the backup and retainng the databases. A week later I checked the backup device again and found that it is still adding database dumps to the file, but it is not overwriting the dumps that are 7 days old. I am told that I need the INIT statement in the dumpdevice command in order for the RETAIN DAYS = command to kick in. However, this did not work before. I can not continue adding to the dump device as it will get too large and take up too much disk space on the server. The only way I can get the device to a reasonable size is to do a backup and initialize the device, which gets rid of all of the older databases. This seems like a catch 22. Is there any way I can make this work. I am running SQL 6.5 on the latest service pack (5a). Any help would be greatly appreciated.
hi,There is facility to do backup of the database. I can use thatfacility to backup the database. But before i do a backup i want tocheck if the disk space available enough to backup that database. Ihave a 22gb database mdf file, when i took a backup of that its only3gb. SO i cannot use the size of mdf file to determine the databasedump file. Is there any facility available to find out the backupdatabase dump size before doing the backup?ThanksNabhonil.
We are running sql 6.5 with Serv. pack 4. All of our Prod. database are backedup weekly via scheduled tasks. On occasion the dump of a particular database will cause 1 of 2 things: 1. The db dump seems to work correctly but stops sql executive service. 2. The db dump seems to work correctly but sends an alert email msg to me indicating that it failed. The msg in the email is "NO Message".
Has anyone experienced either of these 2 problems or have any solutions?
Is anyone using the -DelBkUps option with their maintenance plan on SQL 7.0?? if so is it actually deleting the files? Mine is running but not deleting any of the old backup files.
I would like to run the following SQLMAINT command on my database every 14days:
At 2:00 AM, this command needs to kick off: SQLMAINT.EXE -D cabinet -CkAl -CkCat -UpdSts -RebldIdx 100 -Rpt D:MSSQLLOGcab_maint.rpt
Question: How can I "kill" all users, then place the database in "dbo" mode, run "SQLMAINT", and then release database access back to the public - all in an automated process.
Was wanting to get some feedback on using sqlmaint to produce database backups in a production environment. Especially on databases > 10GB. My preference is not to use it but I'm arguing with another DBA and wanted some outside feedback.
I am a big fan of the backup naming convention used by maintenance plans and the SQLMAINT utility. My problem is that I want to use the same naming convention "db_name_diff_YYYYMMddhhmmss" because of my situation: I'm in California, my databases are in Pennsylvania, I don't have domain access to the servers, I use a scheduling tool (required) that only uses T-SQL or cmd line utilities, and I want the backups named as above. SQLMAINT provides for full and tlog backups, which I have set up, but I want to do diff backups the same way, with the same naming convention. Any ideas?
I run the SQLMAINT.EXE utility nightly to do DBCC checkdb, checkalloc, and checkcat checks on all databases. On one server this check seems to lock up the server displaying error 17832, "unable to read login packets". Once this occurs the only way we can find to get out is to reboot the NT Server. We have observed this behaviour once every two or three weeks. Anyone seen anything similar?
SQL2000 SP3a I would like to generate an HTML output file as a report which should be possible. After creating the maintenance plan using the wizard I used the created syntax and established a second job which now uses the changed syntax - however the job doesnt run. In the log file it just states the basic sqlmaint switches. Current job syntax - SQL uses build in system account.
I have a maintenance plan on a client's site, there's only one databasein the plan and it's set to backup db and log, this job failsconsistently. The one step in the job contains this line:EXECUTE master.dbo.xp_sqlmaint N'-PlanID392B2334-6800-4655-A6C7-5414D9072505 -WriteHistory -VrfyBackup-BkUpMedia DISK -BkUpDB "C:Program FilesMicrosoft SQLServerMSSQLBACKUPDailyBackups" -DelBkUps 4WEEKS -BkExt "BAK"'Which fails if run in QA with the message:Server: Msg 22029, Level 16, State 1, Line 0sqlmaint.exe failed.I've searched for this error and cannot find anything to help.There's no "-S" in the path, it writes absolutely no history on theplan. I set the plan to write history in the log directory on theserver, it wrote nothing to that directory.The database doesn't get backed up but is able to be backed up manuallyto the path specified in the plan. I also have a separate job with aSQL backup statement in it that works OK so SQLAgent is working.
I am using SQL 2000,everyday we are running database backup plan but it shows a error like "Executed as user: NT AUTHORITYSYSTEM. sqlmaint.exe failed. [SQLSTATE 42000] (Error 22029). The step failed." Our database recovery model is also in Full mode.Burt i don't know why its happening with us,before that database maintenance plan was working fine but for last 2 months its not working.
