[Shadow]

SSH tunneling has been extensively discussed in this forum. Please, do a little search first!

[Shadow]

Ususally the error is raised either if a packet is too large or if the connection times out. If you issue “large” INSERT or UPDATE statements (e.g. BLOB, TEXT columns), then you should consider setting max_allowed_packet variable to a larger value. Otherwise connect_timeout, wait_timeout parameters should be tuned in my.ini or my.cnf. Of course, MySql server may be faulty (crashes), but that case is unlikely.

[Shadow]

Sure. Tools -> Database Synchronization Wizard

Choose one-way sync, set source database to the one you want to backup and set the target to another where you want to save the data of the primary db.

[Shadow]

Have you checked the status “dboticketusuario”? It may be corrupted…

I created two tables according to the supplied .sql file, put some test data into them and started syncing. It worked for me!

[Shadow]

What if you tell the job to abort on error and then run it? Did you check sja's log?

[Shadow]

Lots of things could be happening, you should provide us more info, e. g. table structure, sample data, version of target and source servers and SQLyog, anything you can think of!

[Shadow]

Well, in this case you'll have to write a separate app to do the sync for you. It would not be a bad idea though to have a sync option that determines one or more columns and conditions that would drive the two-way sync process!

[Shadow]

mysqladmin is a command line tool that can be run with different parameters, while FLUSH HOSTS is a MySql specific SQL command (at least sort of…). Both do the same thing.

[Shadow]

There must be (or must have been) some network connection problem between you and the machine running MySql server and MySql locked your machine out. FLUSH HOSTS command will reset the error counter and will let your machine connect in case of no further errors.

[Shadow]

You have to create the empty database first, then you'll be able to get it synced to your remote one.

[Shadow]

Currently you cannot do that with sja, but it's a good suggestion. Hope, it makes MAX!

Unfortunately, I can't even think of a workaround using sja, unless the target and source dbs are updated in blocks (e.g.: s. db is updated p.m., while target is updated in the afternoons). In this case WHERE clauses can be used to select which rows to compare!

[Shadow]

That's right, I thought the job file got modified. It didn't…

[Shadow]

Tunneling SQLyog through PHP is not that easy since they do not use the same protocol, PHP uses HTTP, while SQLyog uses MySql's own protocol. A new modul should be built that would translate the requests to http and then interpret the answers. You should try using ssh or any other available tunneling app instead of asking for a HTTP conversion tool!

[Shadow]

And what about the possible bug?

[Shadow]

Interesting. If id is used (it has the same value in both tables), then sync works as expected. If column1 is used, however, the row gets updated despite the fact that the values are the same.

Is it a bug, Ritesh?

[Author]

Posts