I'm following the steps here
https://github.com/robtweed/qewd-starter-kit-yottadb without much success. I know the sweet path is through Docker, but I would rather not have a middleman between the OS and QEWD.
Here is what I have done:
1. Prepared a fresh and clean arm64 Debian 12: Linux development 6.1.0-37-cloud-arm64 #1 SMP Debian 6.1.140-1 (2025-05-22) aarch64
2. Installed the latest YottaDB (r2.02). This works fine; all needed ydb environment variables are sourced and they have the correct values.
3. Installed the latest Node.js (v24.5.0)
4. Installed C++ compiler
5. Cloned the repo mentioned above to the ~/qewd folder
6. Started QEWD up twice, as directed. The first time it went OK, but the second time I got the errors below.
It looks like the workers are expecting YottaDB v1.30 on x86 ?!? What have I missed?
debian@development:~/qewd$ npm start
> qew...@1.0.0 start
> node node_modules/qewd/up/run_native
qewd-client installed
mg-webComponents installed
*** Installation Completed. QEWD will halt ***
*** Please restart QEWD again using "npm start"
debian@development:~/qewd$ npm start
> qew...@1.0.0 start
> node node_modules/qewd/up/run_native
** loading /home/debian/qewd/configuration/config.json
Checking for onWorkerStarted path: /home/debian/qewd/orchestrator/onWorkerStarted.js
Checking for onWorkerStarted path: /home/debian/qewd/onWorkerStarted.js
** results = {
"routes": [],
"config": {
"managementPassword": "secret",
"serverName": "QEWD-Up Server",
"port": 8080,
"poolSize": 2,
"poolPrefork": false,
"idleLimit": "3600000",
"database": {
"type": "dbx"
},
"webServerRootPath": "/home/debian/qewd/www/",
"cors": true,
"bodyParser": false,
"mode": "production",
"sessionDocumentName": "qs",
"qewd_up": true,
"moduleMap": {
"qewd-monitor-adminui": "/home/debian/qewd/qewd-apps/qewd-monitor-adminui"
}
},
"cwd": "/home/debian/qewd",
"startupMode": "normal"
}
config: {
"managementPassword": "secret",
"serverName": "QEWD-Up Server",
"port": 8080,
"poolSize": 2,
"poolPrefork": false,
"idleLimit": "3600000",
"database": {
"type": "dbx"
},
"webServerRootPath": "/home/debian/qewd/www/",
"cors": true,
"bodyParser": false,
"mode": "production",
"sessionDocumentName": "qs",
"qewd_up": true,
"moduleMap": {
"qewd-monitor-adminui": "/home/debian/qewd/qewd-apps/qewd-monitor-adminui"
}
}
routes: []
Double ended queue max length set to 20000
webServerRootPath = /home/debian/qewd/www/
Worker Bootstrap Module file written to node_modules/ewd-qoper8-worker.js
========================================================
ewd-qoper8 is up and running. Max worker pool size: 2
========================================================
========================================================
QEWD.js is listening on port 8080
========================================================
no available workers
sent qoper8-start message to 53914
process.argv[2] = qewd.worker
workerModule: qewd; worker
{
database: { type: 'YottaDB', release: 'r1.30', architecture: 'x86' }
}
Session Garbage Collector has started in worker 53914
Mon, 04 Aug 2025 04:01:45 GMT; master process received response from worker 53914: {"type":"workerProcessStarted","ok":53914}
new worker 53914 started and ready so process queue again
Mon, 04 Aug 2025 04:01:45 GMT; worker 53914 received message: {"type":"ewd-register","application":"qewd-monitor-adminui","jwt":false,"socketId":"2EtdCjG-IzgxDDrcAAAA","ipAddress":"::ffff:xxx.xxx.xxx.xxx"}
#
# Fatal error in , line 0
# Check failed: content_capacity <= capacity.
#
#
#
#FailureMessage Object: 0xffffe4bbf060
----- Native stack trace -----
1: 0x10d5018 [/home/debian/.nvm/versions/node/v24.5.0/bin/node]
2: 0x2a93c40 V8_Fatal(char const*, ...) [/home/debian/.nvm/versions/node/v24.5.0/bin/node]
3: 0x189504c v8::internal::String::WriteUtf8(v8::internal::Isolate*, v8::internal::DirectHandle<v8::internal::String>, char*, unsigned long, v8::base::Flags<v8::internal::String::Utf8EncodingFlag, int, int>, unsigned long*) [/home/debian/.nvm/versions/node/v24.5.0/bin/node]
4: 0x12ecfb0 v8::String::WriteUtf8V2(v8::Isolate*, char*, unsigned long, int, unsigned long*) const [/home/debian/.nvm/versions/node/v24.5.0/bin/node]
5: 0xffffb819e0d4 dbx_write_char8(v8::Isolate*, v8::Local<v8::String>, char*, int, int) [/home/debian/qewd/node_modules/mg-dbx/build/Release/mg-dbx.node]
6: 0xffffb81a5e5c dbx_global_reset(v8::FunctionCallbackInfo<v8::Value> const&, v8::Isolate*, tagDBXCON*, tagDBXMETH*, void*, int, short) [/home/debian/qewd/node_modules/mg-dbx/build/Release/mg-dbx.node]
7: 0xffffb81b8bd4 mglobal::New(v8::FunctionCallbackInfo<v8::Value> const&) [/home/debian/qewd/node_modules/mg-dbx/build/Release/mg-dbx.node]
8: 0x1332a90 [/home/debian/.nvm/versions/node/v24.5.0/bin/node]
9: 0x1332d44 v8::internal::Builtin_HandleApiConstruct(int, unsigned long*, v8::internal::Isolate*) [/home/debian/.nvm/versions/node/v24.5.0/bin/node]
10: 0x204d674 [/home/debian/.nvm/versions/node/v24.5.0/bin/node]
%YDB-F-KILLBYSIGSINFO1, YottaDB process 53914 has been killed by a signal 5 at address 0x0000000002A9E0A0 (vaddr 0x0000000002A9E0A0)
*** master received exit event from worker process 53914
^C*** CTRL & C detected: shutting down gracefully...
No worker processes are running
Master process will now shut down
debian@development:~/qewd$