[DOCS-2330] Our distributed replicaset recommendation in docs don't match w/ support Created: 02/Dec/13 Updated: 30/Oct/23 Resolved: 07/Apr/16 |
|
| Status: | Closed |
| Project: | Documentation |
| Component/s: | manual |
| Affects Version/s: | None |
| Fix Version/s: | Server_Docs_20231030 |
| Type: | Improvement | Priority: | Major - P3 |
| Reporter: | Nicholas Tang | Assignee: | Kay Kim (Inactive) |
| Resolution: | Done | Votes: | 2 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||||||||||||||||||
| Participants: | |||||||||||||||||||||
| Days since reply: | 7 years, 44 weeks, 6 days ago | ||||||||||||||||||||
| Description |
|
When a user asks about setting up geographically diverse replicasets, we recommend using an odd number of data centers so that they can keep a majority live even after one fails. Obviously, for some customers that's not an option, but for 99% of them, sticking an arbiter up in the cloud is actually feasible and is the best answer to their problems. Our docs, however, recommend putting a majority of nodes in a single location, thereby guaranteeing that if that location goes offline, that their replicaset will be read-only (or if it's a sharded environment, completely unusable). Mentioning that as an option for users who absolutely, positively can't have a 3rd location is fine, but it shouldn't be the default recommendation. Doc is here: Thanks, |
| Comments |
| Comment by Githook User [ 07/Apr/16 ] |
|
Author: {u'username': u'kay-kim', u'name': u'kay', u'email': u'kay.kim@10gen.com'}Message: |
| Comment by Githook User [ 07/Apr/16 ] |
|
Author: {u'username': u'kay-kim', u'name': u'kay', u'email': u'kay.kim@10gen.com'}Message: |
| Comment by Githook User [ 07/Apr/16 ] |
|
Author: {u'username': u'kay-kim', u'name': u'kay', u'email': u'kay.kim@10gen.com'}Message: |
| Comment by Githook User [ 07/Apr/16 ] |
|
Author: {u'username': u'kay-kim', u'name': u'kay', u'email': u'kay.kim@10gen.com'}Message: |
| Comment by Githook User [ 07/Apr/16 ] |
|
Author: {u'username': u'kay-kim', u'name': u'kay', u'email': u'kay.kim@10gen.com'}Message: |
| Comment by Githook User [ 07/Apr/16 ] |
|
Author: {u'username': u'kay-kim', u'name': u'kay', u'email': u'kay.kim@10gen.com'}Message: |
| Comment by Githook User [ 07/Apr/16 ] |
|
Author: {u'username': u'kay-kim', u'name': u'kay', u'email': u'kay.kim@10gen.com'}Message: |
| Comment by Githook User [ 07/Apr/16 ] |
|
Author: {u'username': u'kay-kim', u'name': u'kay', u'email': u'kay.kim@10gen.com'}Message: |
| Comment by Githook User [ 06/Apr/16 ] |
|
Author: {u'username': u'kay-kim', u'name': u'kay', u'email': u'kay.kim@10gen.com'}Message: |
| Comment by Alexander Marquardt [ 09/Mar/16 ] |
|
The above recommendations are not totally consistent with our Whitepaper https://www.mongodb.com/collateral/mongodb-multi-data-center-deployments, which gives an example with three data centres that have a 2-2-1 configuration. |
| Comment by Damien Gasparina [ 09/Mar/16 ] |
|
You can find the same kind of recommendation on:
I think it would be best to recommend by default to avoid keeping a majority in one location. This to spare a single point of failure and to allow automatic failover instead of manual failover in case of a DC/general issue. Cheers, |