When using the aggregation mode, the display alias feature display the wrong title: it displays a bogus alias instead of the expected text.

The reason is that the aggregation code is modifying the title of the instance but not its alias.

PR on the way.

According to the postgresql documentation on https://www.postgresql.org/docs/current/static/continuous-archiving.html section 25.3.3.1 you can pass 2 extra parameters to pg_start_backup().

I'm particularly interested in the second parameter which does a fast checkpoint.

Is this a feature which can be added as a command-line parameter to wal-e backup-push eg `wal-e backup-push --fast-chec

URL: https://bugzilla.redhat.com/1379982
Creator: ndevos at redhat
Time: 20160928T10:44:57

Description of problem:
http://review.gluster.org/14701 adds new public functions for gfapi applications that want to receive upcall events. These functions are sparsely documented, which need to improve.

Version-Release number of selected component (if applicable):
mainline (+backports)

In regards to #320, I didnt see any updates in sanoid.conf or sanoid.defaults.conf.

I tried adding zfs_recursion = 1 to my template, but that didnt work.

What is the proper method for enabling atomic recursive snapshots?

Thanks!

The current repmgrd fencing technique (repmgrd invoking the child_nodes_disconnect_command) has 2 problems:

1. It only works when repmgrd is running and failover=automatic
2. It doesn't work if a failover happens when the node where the old primary was running is down at the time of the failover

The repmgrd documentation states (https://repmgr.org/docs/5.0/repmgrd-primary-child-disconnection

I don't want to use current chunks loop for replication (mostly because is still unclear how it works).
What I would like is to set our cluster to replicate as much chunks as possible up to REPLICATION_BANDWIDTH_LIMIT_KBPS

What should I have to set and how ? I think that both REPLICATION_BANDWIDTH_LIMIT_KBPS and chunks loop limits are honored, so I have to set chunk loop to very high value

Hey, I noticed in the MongoShake-Detailed-Documentation.md, at two places, 'Syncer' has been written as 'Synder' twice.

 # 4. Architecture & Data Flow
... 
-Syncer is responsible for fetching data from the source database. If the source is a Mongod or a ReplicaSet, then there is only one **Synder**. If it is a Sharding mode, then there need to be multiple "syncers" connecting to each sh

Hi,
Now that we have nested group in gitlab since v9 it would be very interesting to implement it in gitlab-mirrors.
Would be super useful, as we could mirror several git repositories of a unique project to a unique sub group in a base "gitlab-mirrors" group (wow!)

Cheers

On twitter, Anne Urai asked if it was ok to submit the reproduction of a tutorial. The paper is:

http://times.cs.uiuc.edu/course/410/note/mle.pdf

explanation for non-experts of

• http://persci.mit.edu/pub_pdfs/spatio85.pdf
• http://www.cns.nyu.edu/heegerlab/content/publications/Heeger-JOSA1987.pdf

What do you think ?

TiDB supports online schema change natively while MySQL often leverages tools like gh-ost. For task syncing from MySQL to TiDB, we'd better come up with a solution.

@xialuo1990 could you list ddl stmts gh-ost would execute during schema change?

I'm confused about the example for data migration during code updates. The runMigration example function asserts that this is an upgrade from version 0 to version 1. Shouldn't this be from version 1 to version 2? Or am I completely misunderstanding how migrations work?

Seems like the published docs https://www.2ndquadrant.com/en/resources/pglogical/pglogical-docs/

are not up to date with the info in the readme https://github.com/2ndQuadrant/pglogical

This wasted a lot of time for me because of a couple pieces of missing info. Might be good to merge them, or take one down.

Thanks for a great project!

I want to call External Scripts after swichover/failover the new master and all slaves active . but the "failover-post-scripts" called after the new leader has been setup. can I do this ? how do it ?

Description

在全量迁移时 create table 阶段会执行 count操作，表创建完了，开始迁移数据时还会count。 所以create table阶段的 count操作可否省略掉。

Steps to reproduce the issue:
1.
2.
3.

Describe the results you received:

Describe the results you expected:

Output of ./dtle version:

(paste your output here)

**Additional information you deem important (e.g. issue happens only occasionally