SDL starts as expected by connecting to a specified lite server, however after 5-15 minutes it crashes, it must be related to the new lightwalletd code. I tested this behavior on all available servers.
[error] 20034#20034: *1988 no live upstreams while connecting to upstream, client: 1.2.3.4, server: lite.hush.land, request: "POST /cash.z.wallet.sdk.rpc.CompactTxStreamer/GetLightdInfo HTTP/2.0", upstream: "grpc://localhost", host: "lite.hush.land:443"
SDL starts as expected by connecting to a specified lite server, however after 5-15 minutes it crashes, it must be related to the new lightwalletd code. I tested this behavior on all available servers.
Started testing again to be sure:
| Lite server | Start Time (min) | End time (min) |
| ---------------------------- | ------------- | -------------- |
| `lite.hush.land` | 24 | 34 (crash) |
| `lite.hush.community` | 47 | 32 (no crash) |
| `lite.hushpool.is` | 32 | 33 (crash) |
| `lite.hushpool.is` | 34 | 52 (no crash) |
| `lite2.hushpool.is` | 53 | 13 (no crash) |
| `poop.granitefone.me` | - | - |
| `wtfistheinternet.hush.is` | - | - |
| `lite.myhush.org` | runs `v0.1.1` | - |
Logs:
```
[error] 20034#20034: *1988 connect() failed (111: Connection refused) while connecting to upstream, client: 1.2.3.4, server: lite.hush.land, request: "POST /cash.z.wallet.sdk.rpc.CompactTxStreamer/GetLightdInfo HTTP/2.0", upstream: "grpc://[::1]:9067", host: "lite.hush.land:443"
```
```
[error] 20034#20034: *1988 no live upstreams while connecting to upstream, client: 1.2.3.4, server: lite.hush.land, request: "POST /cash.z.wallet.sdk.rpc.CompactTxStreamer/GetLightdInfo HTTP/2.0", upstream: "grpc://localhost", host: "lite.hush.land:443"
```
Those errors are nginx specific, I removed listen host:443 ssl http2; and no issues so far. I can't explain why SDL doesn't crash anymore using other servers, I think it was a false alarm, moved lite.myhush.org to the latest code again and closing.
Those errors are nginx specific, I removed `listen host:443 ssl http2;` and no issues so far. I can't explain why SDL doesn't crash anymore using other servers, I think it was a false alarm, moved `lite.myhush.org` to the latest code again and closing.
SDL starts as expected by connecting to a specified lite server, however after 5-15 minutes it crashes, it must be related to the new lightwalletd code. I tested this behavior on all available servers.
Started testing again to be sure:
lite.hush.landlite.hush.communitylite.hushpool.islite.hushpool.islite2.hushpool.ispoop.granitefone.mewtfistheinternet.hush.islite.myhush.orgv0.1.1Logs:
In progress.
@onryo are you able to get a backtrace from a coredump?
Those errors are nginx specific, I removed
listen host:443 ssl http2;and no issues so far. I can't explain why SDL doesn't crash anymore using other servers, I think it was a false alarm, movedlite.myhush.orgto the latest code again and closing.