[jira] [Commented] (FINERACT-982) Completely ditch use of Drizzle JDBC Driver after all
[ https://issues.apache.org/jira/browse/FINERACT-982?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17385757#comment-17385757 ] VICTOR MANUEL ROMERO RODRIGUEZ commented on FINERACT-982: - We have done some testing with the collation and MariaDB, the PR is https://github.com/apache/fineract/pull/1799 > Completely ditch use of Drizzle JDBC Driver after all > - > > Key: FINERACT-982 > URL: https://issues.apache.org/jira/browse/FINERACT-982 > Project: Apache Fineract > Issue Type: Improvement >Reporter: Michael Vorburger >Priority: Blocker > Labels: scalability > Fix For: 1.6.0 > > > Fineract's use of the un-maintained Drizzle JDBC driver continues to cause > confusion and pains like FINERACT-980, and note e.g. the proposed removal of > the alternative MySQL JDBC driver in > [https://github.com/apache/fineract/pull/887.] > Some of the background to this is recorded e.g. in FINCN-26, FINERACT-761 and > LEGAL-462. > LEGAL-462 has clarified that the Fineract ZIP distribution must include > Drizzle and cannot distribute the MariaDB or MySQL JDBC clients. We CAN, and > currently (given the confusion in FINERACT-980 apparently do?!) use it for > tests. > What if to reduce variability we just removed that old Drizzle JDBC driver > after all? We could run our tests using (preferably) the very well maintained > MariaDB JDBC driver client (the actual DB server is a totally separate > discussion, see FINERACT-896). We would (have to) distribute the ZIP for > download without a JDBC driver, and just some documentation inviting users to > DL and add one. > But the exact situation about distributing in a Docker container image isn't > clear, to me... > FYI [~awasum], [~ptuomola], [~xurror] -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (FINERACT-982) Completely ditch use of Drizzle JDBC Driver after all
[ https://issues.apache.org/jira/browse/FINERACT-982?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17384677#comment-17384677 ] VICTOR MANUEL ROMERO RODRIGUEZ commented on FINERACT-982: - Hellow, Could you please check at [https://github.com/apache/fineract/pull/1798] The Drizzle Driver has been changed to MariaDB and the charset and collation also has been replaced to utf8mb4 and utf8_unicode_ci > Completely ditch use of Drizzle JDBC Driver after all > - > > Key: FINERACT-982 > URL: https://issues.apache.org/jira/browse/FINERACT-982 > Project: Apache Fineract > Issue Type: Improvement >Reporter: Michael Vorburger >Priority: Blocker > Labels: scalability > Fix For: 1.6.0 > > > Fineract's use of the un-maintained Drizzle JDBC driver continues to cause > confusion and pains like FINERACT-980, and note e.g. the proposed removal of > the alternative MySQL JDBC driver in > [https://github.com/apache/fineract/pull/887.] > Some of the background to this is recorded e.g. in FINCN-26, FINERACT-761 and > LEGAL-462. > LEGAL-462 has clarified that the Fineract ZIP distribution must include > Drizzle and cannot distribute the MariaDB or MySQL JDBC clients. We CAN, and > currently (given the confusion in FINERACT-980 apparently do?!) use it for > tests. > What if to reduce variability we just removed that old Drizzle JDBC driver > after all? We could run our tests using (preferably) the very well maintained > MariaDB JDBC driver client (the actual DB server is a totally separate > discussion, see FINERACT-896). We would (have to) distribute the ZIP for > download without a JDBC driver, and just some documentation inviting users to > DL and add one. > But the exact situation about distributing in a Docker container image isn't > clear, to me... > FYI [~awasum], [~ptuomola], [~xurror] -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Comment Edited] (FINERACT-982) Completely ditch use of Drizzle JDBC Driver after all
[ https://issues.apache.org/jira/browse/FINERACT-982?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17384677#comment-17384677 ] VICTOR MANUEL ROMERO RODRIGUEZ edited comment on FINERACT-982 at 7/21/21, 6:11 AM: --- Hello, Could you please take a look at [https://github.com/apache/fineract/pull/1798] ? The Drizzle Driver has been changed to MariaDB and the charset and collation also has been replaced to utf8mb4 and utf8_unicode_ci was (Author: vicromero): Hellow, Could you please check at [https://github.com/apache/fineract/pull/1798] The Drizzle Driver has been changed to MariaDB and the charset and collation also has been replaced to utf8mb4 and utf8_unicode_ci > Completely ditch use of Drizzle JDBC Driver after all > - > > Key: FINERACT-982 > URL: https://issues.apache.org/jira/browse/FINERACT-982 > Project: Apache Fineract > Issue Type: Improvement >Reporter: Michael Vorburger >Priority: Blocker > Labels: scalability > Fix For: 1.6.0 > > > Fineract's use of the un-maintained Drizzle JDBC driver continues to cause > confusion and pains like FINERACT-980, and note e.g. the proposed removal of > the alternative MySQL JDBC driver in > [https://github.com/apache/fineract/pull/887.] > Some of the background to this is recorded e.g. in FINCN-26, FINERACT-761 and > LEGAL-462. > LEGAL-462 has clarified that the Fineract ZIP distribution must include > Drizzle and cannot distribute the MariaDB or MySQL JDBC clients. We CAN, and > currently (given the confusion in FINERACT-980 apparently do?!) use it for > tests. > What if to reduce variability we just removed that old Drizzle JDBC driver > after all? We could run our tests using (preferably) the very well maintained > MariaDB JDBC driver client (the actual DB server is a totally separate > discussion, see FINERACT-896). We would (have to) distribute the ZIP for > download without a JDBC driver, and just some documentation inviting users to > DL and add one. > But the exact situation about distributing in a Docker container image isn't > clear, to me... > FYI [~awasum], [~ptuomola], [~xurror] -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (FINCN-189) Add /health endpoint to each micro service
[ https://issues.apache.org/jira/browse/FINCN-189?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17007232#comment-17007232 ] VICTOR MANUEL ROMERO RODRIGUEZ commented on FINCN-189: -- Hi Juhan & Fineract folks, I think the logging and monitoring should be done in two parts: # Check for the infrastructure availability - health check/readiness # Check for the health and gather information of each Microservice for feeding the logging and monitoring applications like Prometheus, ELK, Splunk or APM like AppDynamics. For each Docker container we have done a quick solution for checking if the availability of each service, not only the TCP port checking but for the database readiness or the backend on the other side of the middleware messaging. Initially we have done the improvement for point 1 here: [https://github.com/fintecheando/fineract-cn-docker-compose] [https://github.com/itzamnamx/fineract-cn-lang-security-docker] [https://github.com/itzamnamx/fineract-cn-provisioner] [https://github.com/fintecheando/fineract-cn-identity] What we have done is using the docker segmentation of the services and the dependency verification for the startup. We have tested this in our local dev and works. If this is Ok, please assign me this ticket so then I will send the PRs to the Fineract repositories. For point 2 we have used Spring boot actuator for accomplish the task but is required to have the Logging and Monitoring tools enabled, we prefer to use ELK or Prometheus. Let me know your thoughts. Regards Victor Romero > Add /health endpoint to each micro service > -- > > Key: FINCN-189 > URL: https://issues.apache.org/jira/browse/FINCN-189 > Project: Fineract Cloud Native > Issue Type: Improvement >Reporter: Juhan Aasaru >Priority: Major > Labels: volunteer > > In order to run Fineract-CN micro services in production we need to know > about each service if it is running or failing. > For this we need to create /health endpoints to each micro service. If anyone > turns to this endpoint the program code checks if it can access Postgres and > Cassandra and if yes it responds back "UP", otherwise "DOWN". > I recommend starting with Identity micro service because this micro service > doens't depend on any other micro service (to run it you only need to have > Cassandra and Postgres running). > -- This message was sent by Atlassian Jira (v8.3.4#803005)
