pipeline fail after implementing hikari connection pooling

Asmaa Ammar

We've recently implemented hikari connection pooling to manage database connections, the tests are passing locally but in the pipeline, it is failing with error message: 
<span>unexpected EOF on client connection with an open transaction</span>

 

 

<span>PostgreSQL init process complete; ready for start up.</span>

49

50

<span>2023-10-18 04:22:03.143 UTC [1] LOG: starting PostgreSQL 16.0 (Debian 16.0-1.pgdg120+1) on x86_64-pc-linux-gnu, compiled by gcc (Debian 12.2.0-14) 12.2.0, 64-bit</span>

51

<span>2023-10-18 04:22:03.143 UTC [1] LOG: listening on IPv4 address "0.0.0.0", port 5432</span>

52

<span>2023-10-18 04:22:03.143 UTC [1] LOG: listening on IPv6 address "::", port 5432</span>

53

<span>2023-10-18 04:22:03.144 UTC [1] LOG: listening on Unix socket "/var/run/postgresql/.s.PGSQL.5432"</span>

54

<span>2023-10-18 04:22:03.147 UTC [62] LOG: database system was shut down at 2023-10-18 04:22:03 UTC</span>

55

<span>2023-10-18 04:22:03.152 UTC [1] LOG: database system is ready to accept connections</span>

56

<span>2023-10-18 04:27:03.246 UTC [60] LOG: checkpoint starting: time</span>

57

<span>2023-10-18 04:27:07.625 UTC [60] LOG: checkpoint complete: wrote 44 buffers (0.3%); 0 WAL file(s) added, 0 removed, 0 recycled; write=4.115 s, sync=0.123 s, total=4.380 s; sync files=11, longest=0.123 s, average=0.012 s; distance=260 kB, estimate=260 kB; lsn=0/152BEA0, redo lsn=0/152BE68</span>

58

<span>2023-10-18 04:28:55.452 UTC [85] ERROR: relation "public.databasechangeloglock" does not exist at character 22</span>

59

<span>2023-10-18 04:28:55.452 UTC [85] STATEMENT: SELECT COUNT(*) FROM public.databasechangeloglock</span>

60

<span>2023-10-18 04:28:55.452 UTC [86] ERROR: relation "public.databasechangeloglock" does not exist at character 22</span>

61

<span>2023-10-18 04:28:55.452 UTC [86] STATEMENT: SELECT COUNT(*) FROM public.databasechangeloglock</span>

62

<span>2023-10-18 04:28:55.452 UTC [83] ERROR: relation "public.databasechangeloglock" does not exist at character 22</span>

63

<span>2023-10-18 04:28:55.452 UTC [83] STATEMENT: SELECT COUNT(*) FROM public.databasechangeloglock</span>

64

<span>2023-10-18 04:28:55.452 UTC [84] ERROR: relation "public.databasechangeloglock" does not exist at character 22</span>

65

<span>2023-10-18 04:28:55.452 UTC [84] STATEMENT: SELECT COUNT(*) FROM public.databasechangeloglock</span>

66

<span>2023-10-18 04:28:55.860 UTC [60] LOG: checkpoint starting: immediate force wait</span>

67

<span>2023-10-18 04:28:55.880 UTC [60] LOG: checkpoint complete: wrote 932 buffers (5.7%); 0 WAL file(s) added, 0 removed, 1 recycled; write=0.017 s, sync=0.003 s, total=0.021 s; sync files=302, longest=0.001 s, average=0.001 s; distance=17101 kB, estimate=17101 kB; lsn=0/25DF500, redo lsn=0/25DF4C0</span>

68

<span>2023-10-18 04:28:55.982 UTC [85] LOG: unexpected EOF on client connection with an open transaction</span>

Asmaa Ammar

Dear Theodora,

 

Thank you so much for your response and explanation. It turned out to be memory allocation issue in the pipeline. Reducing docker memory to 512MB solved my issue.