[COMPASS-3238] SSH tunnel: topology was destroyed Created: 06/Nov/18 Updated: 12/Feb/20 Resolved: 12/Feb/20 |
|
| Status: | Closed |
| Project: | Compass |
| Component/s: | Connectivity |
| Affects Version/s: | 1.15.4 |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | airs0urce | Assignee: | Lucas Hrabovsky (Inactive) |
| Resolution: | Done | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Environment: |
MacOS 10.14.1, Macbook Pro 15 2018. Connected to mongoDB server via SSH |
||
| Attachments: |
|
||||
| Issue Links: |
|
||||
| Sprint: | Iteration Xerus | ||||
| Description |
|
I connected to mongo server via SSH. When I'm not active for some time (about 15 min) and then try to run some query I get "topology was destroyed" error. Looks like it happens everywhere, when I run basic query or when run aggregation pipeline. I attached screenshot to show you how I see it. |
| Comments |
| Comment by Massimiliano Marcon [ 12/Feb/20 ] |
|
This is gone with the most recent versions of the node driver. |
| Comment by Lucas Hrabovsky (Inactive) [ 13/Nov/19 ] |
|
|
| Comment by Brian Blevins [ 05/Jan/19 ] |
|
If the topology was destroyed error messages while using Compass over an SSH tunnel are caused by idle timeouts, then a workaround would be adjusting the /etc/ssh/sshd_config setting in the sshd server to prevent the idle timeouts. Specifically, set or reduce the ClientAliveInterval to a value between 60 and 120 (seconds). Then restart the sshd and reconnect with Compass. Many firewall configurations will forget (drop) connections with no traffic after some timeout interval, often between 5 minutes and 15 minutes. By configuring the connection to have keep-alive packets more frequent that the firewall timeout, the connection can be held open. |
| Comment by airs0urce [ 07/Nov/18 ] |
|
Here is screenshot from another page:
|