caching touch() queries and re-using the plans causing segfault #6560
Description
To reproduce
CREATE TABLE t1 AS (
SELECT generate_series as t,
(generate_series % 123456)::int as i
FROM generate_series('2025-12-25', '2025-12-26', '1s')
) PARTITION BY DAY WAL;
select touch(select * from t1);
-- works fine
alter table t1 add column f float;
select touch(select * from t1);
-- cached query plan cannot be used because table schema has changed [table=t1', expectedTableId=60408, actualTableId=60408, expectedMetadataVersion=0, actualMetadataVersion=1]
-- get an error, and result, but still ok
select touch(select * from t1);
-- Invalid JSON response from the server: SyntaxError: JSON.parse: unexpected end of data at line 1 column 1 of the JSON data
-- server segfaulted and process killedWorse, this persists even after restart. If you try to run select touch(select * from t1); again, it will crash.
server log:
2025-12-18T17:26:06.509789Z I i.q.g.e.QueryProgress exe [id=39, sql=`select touch(select * from t1)`, principal=admin, cache=false, jit=false]
#
# A fatal error has been detected by the Java Runtime Environment:
#
# SIGSEGV (0xb) at pc=0x000000011283a8a0, pid=59295, tid=40963
#
# JRE version: OpenJDK Runtime Environment Corretto-17.0.11.9.1 (17.0.11+9) (build 17.0.11+9-LTS)
# Java VM: OpenJDK 64-Bit Server VM Corretto-17.0.11.9.1 (17.0.11+9-LTS, mixed mode, sharing, tiered, compressed oops, compressed class ptrs, g1 gc, bsd-aarch64)
# Problematic frame:
# J 1949 c2 sun.misc.Unsafe.getByte(J)B [email protected] (8 bytes) @ 0x000000011283a8a0 [0x000000011283a880+0x0000000000000020]
#
# No core dump will be written. Core dumps have been disabled. To enable core dumping, try "ulimit -c unlimited" before starting Java again
#
# An error report file with more information is saved as:
# /Users/nick/IdeaProjects/questdb-4/hs_err_pid59295.log
#
# If you would like to submit a bug report, please visit:
# https://github.com/corretto/corretto-17/issues/
stack trace:
Current thread (0x000000013e808e00): JavaThread "shared-network_2" [_thread_in_Java, id=40963, stack(0x0000000176b68000,0x0000000176d6b000)]
Stack: [0x0000000176b68000,0x0000000176d6b000], sp=0x0000000176d69ad0, free space=2054k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native code)
J 1949 c2 sun.misc.Unsafe.getByte(J)B [email protected] (8 bytes) @ 0x000000011283a8a0 [0x000000011283a880+0x0000000000000020]
j io.questdb.griffin.engine.functions.table.TouchTableFunctionFactory$TouchTableFunc.touchMemory(JJJ)J+30 [email protected]
j io.questdb.griffin.engine.functions.table.TouchTableFunctionFactory$TouchTableFunc.touchTable()V+104 [email protected]
j io.questdb.griffin.engine.functions.table.TouchTableFunctionFactory$TouchTableFunc.getStr(Lio/questdb/std/str/Utf16Sink;)V+1 [email protected]
j io.questdb.griffin.engine.functions.table.TouchTableFunctionFactory$TouchTableFunc.getStrA(Lio/questdb/cairo/sql/Record;)Ljava/lang/CharSequence;+12 [email protected]
j io.questdb.cairo.sql.VirtualFunctionRecord.getStrA(I)Ljava/lang/CharSequence;+9 [email protected]
J 3926 c2 io.questdb.cutlass.http.processors.JsonQueryProcessorState.doQueryRecord(Lio/questdb/cutlass/http/HttpChunkedResponse;I)V [email protected] (814 bytes) @ 0x00000001129ae170 [0x00000001129ad5c0+0x0000000000000bb0]
j io.questdb.cutlass.http.processors.JsonQueryProcessorState.doRecordFetchLoop(Lio/questdb/cutlass/http/HttpChunkedResponse;I)V+8 [email protected]
j io.questdb.cutlass.http.processors.JsonQueryProcessorState.onResumeSendRecordsLoop(Lio/questdb/cutlass/http/HttpChunkedResponse;I)V+10 [email protected]
j io.questdb.cutlass.http.processors.JsonQueryProcessorState.onResumeQueryPrefix(Lio/questdb/cutlass/http/HttpChunkedResponse;I)V+22 [email protected]
j io.questdb.cutlass.http.processors.JsonQueryProcessorState.onResumeSetupFirstRecord(Lio/questdb/cutlass/http/HttpChunkedResponse;I)V+27 [email protected]
j io.questdb.cutlass.http.processors.JsonQueryProcessorState$$Lambda$584+0x000000040166ae10.onResume(Lio/questdb/cutlass/http/HttpChunkedResponse;I)V+6 [email protected]
j io.questdb.cutlass.http.processors.JsonQueryProcessorState.resume(Lio/questdb/cutlass/http/HttpChunkedResponse;)V+19 [email protected]
j io.questdb.cutlass.http.processors.JsonQueryProcessor.doResumeSend(Lio/questdb/cutlass/http/processors/JsonQueryProcessorState;Lio/questdb/cutlass/http/HttpConnectionContext;)V+37 [email protected]
j io.questdb.cutlass.http.processors.JsonQueryProcessor.executeSelect0(Lio/questdb/cutlass/http/processors/JsonQueryProcessorState;Lio/questdb/cairo/sql/RecordCursorFactory;Z)V+59 [email protected]
j io.questdb.cutlass.http.processors.JsonQueryProcessor.executeSelect(Lio/questdb/cutlass/http/processors/JsonQueryProcessorState;Lio/questdb/cairo/sql/RecordCursorFactory;)V+4 [email protected]
j io.questdb.cutlass.http.processors.JsonQueryProcessor.executeNewSelect(Lio/questdb/cutlass/http/processors/JsonQueryProcessorState;Lio/questdb/griffin/CompiledQuery;Ljava/lang/CharSequence;)V+8 [email protected]
j io.questdb.cutlass.http.processors.JsonQueryProcessor$$Lambda$495+0x000000040163f030.execute(Lio/questdb/cutlass/http/processors/JsonQueryProcessorState;Lio/questdb/griffin/CompiledQuery;Ljava/lang/CharSequence;)V+7 [email protected]
j io.questdb.cutlass.http.processors.JsonQueryProcessor.compileAndExecuteQuery(Lio/questdb/cutlass/http/processors/JsonQueryProcessorState;)V+108 [email protected]
j io.questdb.cutlass.http.processors.JsonQueryProcessor.execute0(Lio/questdb/cutlass/http/processors/JsonQueryProcessorState;)V+189 [email protected]
j io.questdb.cutlass.http.processors.JsonQueryProcessor.onRequestComplete(Lio/questdb/cutlass/http/HttpConnectionContext;)V+69 [email protected]
j io.questdb.cutlass.http.HttpConnectionContext.handleClientRecv(Lio/questdb/cutlass/http/HttpRequestProcessorSelector;Lio/questdb/cutlass/http/RescheduleContext;)Z+789 [email protected]
j io.questdb.cutlass.http.HttpConnectionContext.handleClientOperation(ILio/questdb/cutlass/http/HttpRequestProcessorSelector;Lio/questdb/cutlass/http/RescheduleContext;)Z+39 [email protected]
j io.questdb.cutlass.http.HttpServer.handleClientOperation(Lio/questdb/cutlass/http/HttpConnectionContext;ILio/questdb/cutlass/http/HttpRequestProcessorSelector;Lio/questdb/cutlass/http/WaitProcessor;Lio/questdb/network/IODispatcher;)Z+5 [email protected]
j io.questdb.cutlass.http.HttpServer$1.lambda$$0(ILio/questdb/cutlass/http/HttpConnectionContext;Lio/questdb/network/IODispatcher;)Z+18 [email protected]
j io.questdb.cutlass.http.HttpServer$1$$Lambda$489+0x0000000401638670.onRequest(ILio/questdb/network/IOContext;Lio/questdb/network/IODispatcher;)Z+10 [email protected]
J 3688 c2 io.questdb.cutlass.http.HttpServer$1.run(ILio/questdb/mp/Job$RunStatus;)Z [email protected] (36 bytes) @ 0x0000000112964510 [0x0000000112964100+0x0000000000000410]
J 3514% c2 io.questdb.mp.Worker.run()V [email protected] (974 bytes) @ 0x000000011290ab50 [0x000000011290aa80+0x00000000000000d0]
v ~StubRoutines::call_stub
V [libjvm.dylib+0x48e248] JavaCalls::call_helper(JavaValue*, methodHandle const&, JavaCallArguments*, JavaThread*)+0x38c
V [libjvm.dylib+0x48d214] JavaCalls::call_virtual(JavaValue*, Klass*, Symbol*, Symbol*, JavaCallArguments*, JavaThread*)+0x11c
V [libjvm.dylib+0x48d2e0] JavaCalls::call_virtual(JavaValue*, Handle, Klass*, Symbol*, Symbol*, JavaThread*)+0x64
V [libjvm.dylib+0x5477b8] thread_entry(JavaThread*, JavaThread*)+0xc4
V [libjvm.dylib+0x9d6af4] JavaThread::thread_main_inner()+0x150
V [libjvm.dylib+0x9d50fc] Thread::call_run()+0xcc
V [libjvm.dylib+0x7f9280] thread_native_entry(Thread*)+0x158
C [libsystem_pthread.dylib+0x6f94] _pthread_start+0x88
siginfo: si_signo: 11 (SIGSEGV), si_code: 2 (SEGV_ACCERR), si_addr: 0x0000000000000000
QuestDB version:
9.2.2+
OS, in case of Docker specify Docker and the Host OS:
MacOS
File System, in case of Docker specify Host File System:
APFS
Full Name:
Nick Woolmer
Affiliation:
QuestDB
Have you followed Linux, MacOs kernel configuration steps to increase Maximum open files and Maximum virtual memory areas limit?
- Yes, I have
Additional context
No response