Activity for Sladjan Vasic
-
Sladjan Vasic posted a comment on discussion Help
We have a router which routes data from "codelist" tables to client nodes based on table specific rules. In past this router was of type default, which meant that all "codelist" data should be routed to all target nodes of targeted node group. Recently we extended database model in a way, that some of the "codelist" tables became "region" specific. These region specific rules are written in a properties table, for each "codelist" table. Based on these rules, router would decide if data from specific...
-
Hello, Is there a way to get a reference to a "table name" of data, that is beeing routed? ie in subselect type I would need somthing like c.external_id in (select STORE_ID from XYZ where column = :COLUMN and :TABLE_NAME like 'XX%')
-
I tried to repeat this problem. I restored database and upgraded symmetricds as in previous try. First it created temp table, then after restart it did move everything back to original table. It didn't make sense to me as I did everything the same as first time, so I retried the whole process again. Now it created the same problem. Temp table was not removed, and all bacthes were left in temp table, while original table was empty. This indeed is strange. I think this may be due to very large table....
-
Yes it is created, but the data from a temp table is not moved back to a "new" table. As a "new", I mean the same SYM_OUTGOING_BATCH table that was altered with some column changes. After restarting SymmetricDS, new batches are creating as usual, but I'm not sure if the batches from a temp SYM_OUTGOING_BATCH_ table, whose status is still in "NE" be sent to nodes, or will these unrouted data changes be lost?
-
We have a problem when registering node. Only one table gets stuck when trying to send batch from server to client. It happens only on reload channel. On the client side we can see error: java.net.SocketTimeoutException: Read timed out at java.net.SocketInputStream.socketRead0(Native Method) at java.net.SocketInputStream.socketRead(Unknown Source) at java.net.SocketInputStream.read(Unknown Source) at java.net.SocketInputStream.read(Unknown Source) at java.io.BufferedInputStream.fill(Unknown Source)...
-
Hi, Recently on test environment, we upgraded symmetricds version from 3.8.30 to 3.9.9 ... We saw that table SYM_OUTGOING_BATCH was (due to uncompatible model changes) renamed to SYM_OUTGOING_BATCH_ (with underscore at the end) New table SYM_OUTGOING_BATCH that was created was empty after upgrade. Will symmetric route data that has not yet been routed to nodes, but was moved to SYM_OUTGOING_BATCH_ table after this upgrade?
-
Hi, Recently on test environment, we upgraded symmetricds version from 3.8.30 to 3.9.X ... We saw that table SYM_OUTGOING_BATCH was (due to uncompatible model changes) renamed to SYM_OUTGOING_BATCH_ (with underscore at the end) New table SYM_OUTGOING_BATCH that was created was empty after upgrade. Will symmetric route data that has not yet been routed to nodes, but was moved to SYM_OUTGOING_BATCH_ table after this upgrade?
-
Hi, Recently on test environment, we upgraded symmetricds version from 3.8.30 to 3.9.X ... We saw that table SYM_OUTGOING_BATCH was (due to uncompatible model changes) renamed to SYM_OUTGOING_BATCH_ (with underscore at the end) New table SYM_OUTGOING_BATCH that is beeing created is empty after upgrade. Will symmetric route data that has not yet been routed, but was moved to SYM_OUTGOING_BATCH_ table?
-
Anyone? Purge job is obviously not purging batches on heartbeat channel. I think this is related to issue https://www.symmetricds.org/issues/view.php?id=3174 but we are using 3.8.30 version, and this bug should be resolved in 3.8.27 Any workaround?
-
Thank you Klementina and Wayne, you cleared a lot of things to me, so I could dig into analyzing a problem with a proper information. I think I identified a problem. I have a lot of batches in status 'OK' with node_ids that do not exists in sym_node table. Some of those batches on our channels are older than 5 months (since day when they were registered, but never used later and deleted from sym_node tables) -- count batches by node_id with status 'OK' not in existing sym_node_host nodes (left used...
-
Thank you Klementina and Wayne, you cleared a lot of things to me, so I could dig into analyzing a problem with a proper information. I think I identified a problem. I have a lot of batches in status 'OK' with node_ids that do not exists in sym_node table. Some of those batches on our channels are older than 5 months (since day when they were registered, but never used later and deleted from sym_node tables) -- count batches by node_id with status 'OK' not in existing sym_node_host nodes (left used...
-
Thank you Klementina and Wayne, you cleared a lot of things to me, so I could dig into analyzing a problem with a proper information. I think I identified a problem. I have a lot of batches in status 'OK' with node_ids that do not exists in sym_node table. Some of those batches on our channels are older than 5 months (since day when they were registered, but never used later and deleted from sym_node tables) -- count batches by node_id with status 'OK' not in existing sym_node_host nodes (left used...
-
Thank you Klementina and Wayne, you cleared a lot of things to me, so I could dig into analyzing a problem with a proper information. I think I identified a problem. I have a lot of batches in status 'OK' with node_ids that do not exists in sym_node table. Some of those batches on our channels are older than 5 months (since day when they were registered, but never used later and deleted from sym_node tables) -- count batches by node_id with status 'OK' not in existing sym_node_host nodes (left used...
-
Thank you Klementina and Wayne, you cleared a lot of things to me, so I could dig into analyzing a problem with a proper information. I think I identified a problem. I have a lot of batches in status 'OK' with node_ids that do not exists in sym_node table. Some of those batches on our channels are older than 5 months (since day when they were registered, but never used later and deleted from sym_node tables) -- count batches by node_id with status 'OK' not in existing sym_node_host nodes (left used...
-
Thank you Klementina and Wayne, you cleared a lot of things to me, so I could dig into analyzing a problem with a proper information. I think I identified a problem. I have a lot of batches in status 'OK' with node_ids that do not exists in sym_node table. -- count batches by node_id with status 'OK' not in existing sym_node_host nodes (left used just to hide sensitive data) SELECT LEFT(NODE_ID, 5) AS NODE, CNT FROM ( SELECT NODE_ID, COUNT(BATCH_ID) AS CNT FROM CASHREGISTER.SYM_OUTGOING_BATCH WHERE...
-
Klementina thanks for response. I must have overlooked it in mail, so I'm responding little late, but hopefully not too late :) I can see purge job running regularly, but It's not clearing the data on one engine. I understand that it should not clear data for offline nodes, but that is not the case here. It's not clearing data even for nodes that are online (at least not for all nodes). One more clue is that in symmetric folder (tmp/engineName/outgoing) there are many (70GB) files with ".done" extension....
-
Hello, Our configuration is working with 2 server nodes (one for test database and one for production database). Every engine is communicating with many (up to 150) client nodes. On a server we are running DB2 database and on a client side MySQL database. I'd like to know how can I reduce table data in theese tables (sym_data and sym_data_event) and why they aren't cleared during the purge job? Inside those tables we have data that is older (based on create_time) than 5 months, even for nodes that...
-
btw ... this version (3.8.30) saved the day :) We had some serious problems with connection timeouts because of some unacknowledged batches which were stuck on a client side (even though data was successfully synchronized to server) and even ran out of memory because of that on a brand new windows surface pro 4. For now it seem like this bug is fixed with the latest version of symmetric server. All missing batches were successfully acknowledged and for now we don't have any timeout or memory problems...
-
btw ... this version (3.8.30) saved the day :) We had some serious problems with connection timeouts because of some unacknowledged batches which were stuck on a client side (even though data was successfully synchronized to server) and even ran out of memory because of that on a brand new windows surface pro 4. For now it seem like this bug is fixed with the latest version of symmetric server. All missing batches were successfully acknowledged and for now we don't have any timeout or memory problems...
-
Sorry ... everything is ok. You can delete this topic.
-
It seems that last version 3.8.30 is not yet available on maven repository (http://maven.jumpmind.com/repo/) When can we expect it to be available? Sladjan
1
