Odd issue when going through tutorial - copy csv fails to copy to table [solved]

[webventurer]

Setup: Vultr Cloud Compute (VC2) server, Ubuntu 18.04.1 LTS, Postgres 10.5, timescaledb extension 0.12.1

hi guys - I was running through the tutorial here:

https://blog.timescale.com/analyzing-ethereum-bitcoin-and-1200-cryptocurrencies-using-postgresql-downloading-the-dataset-a1bbc2d4d992

Ran this command (as per tutorial):
crypto@oracle:~/tmp/crypto_data$ PGPASSWORD=hidden psql -U crypto -d crypto_data_test -h localhost -p 5433 -c "\COPY crypto_prices FROM crypto_prices.csv CSV"
server closed the connection unexpectedly
This probably means the server terminated abnormally
before or while processing the request.

Received these errors in /var/log/postgresql/postgres-10-main.log:
2018-09-20 15:22:31.757 UTC [3299] DETAIL: The postmaster has commanded this server process to roll back the current transaction and exit, because another server process exited abnormally and possibly corrupted shared memory.
2018-09-20 15:22:31.757 UTC [3299] HINT: In a moment you should be able to reconnect to the database and repeat your command.
2018-09-20 15:22:31.765 UTC [1902] LOG: all server processes terminated; reinitializing
2018-09-20 15:22:31.790 UTC [3453] LOG: database system was interrupted; last known up at 2018-09-20 15:16:05 UTC
2018-09-20 15:22:32.584 UTC [3453] LOG: database system was not properly shut down; automatic recovery in progress
2018-09-20 15:22:32.590 UTC [3453] LOG: redo starts at 0/1F400050
2018-09-20 15:22:37.965 UTC [3453] LOG: redo done at 0/337FFF58
2018-09-20 15:22:37.966 UTC [3453] LOG: last completed transaction was at log time 2018-09-20 15:22:10.774633+00
2018-09-20 15:22:38.344 UTC [1902] LOG: database system is ready to accept connections
2018-09-20 15:22:38.344 UTC [3459] LOG: TimescaleDB background worker launcher connected to shared catalogs (edited)
Vultr cloud server, Ubuntu 18.04.1 LTS, Postgres 10.5, timescaledb extension 0.12.1

Note: Everytime I run it I get a different error:
ERROR: index row requires 534432 bytes, maximum size is 8191
CONTEXT: COPY crypto_prices, line 402850: "11/14/2015 19:00,1.10E-06,2.20E-06,1.10E-06,2.20E-06,35751.04,0.0777,SLM"

And on another run another different (but related) error:
ERROR: index row requires 534432 bytes, maximum size is 8191
CONTEXT: COPY crypto_prices, line 520255: "3/16/2015 20:00,4.51E-05,5.00E-05,4.41E-05,4.42E-05,0,0,TAG"

Any ideas what happened?

[webventurer]

And another one:
2018-09-20 16:32:17.573 UTC [4227] crypto@crypto_data_test ERROR: index row requires 534432 bytes, maximum size is 8191
2018-09-20 16:32:17.573 UTC [4227] crypto@crypto_data_test CONTEXT: COPY crypto_prices, line 518124: "3/22/2015 20:00,2.20E-05,2.40E-05,2.20E-05,2.39E-05,12763.37,0.2812,MAX"
2018-09-20 16:32:17.573 UTC [4227] crypto@crypto_data_test STATEMENT: COPY crypto_prices FROM STDIN CSV
2018-09-20 16:32:24.053 UTC [4231] WARNING: relation "pg_attribute" page 465 is uninitialized --- fixing

Seems to get a different error each time I run it

[webventurer]

Issue resolved

I ran the pgtune (https://pgtune.leopard.in.ua/#/) which generated these parameters for my instance:

DB Version: 10

OS Type: linux

DB Type: dw

Total Memory (RAM): 32 GB

CPUs num: 8

Connections num: 10

Data Storage: ssd

max_connections = 10
shared_buffers = 8GB
effective_cache_size = 24GB
maintenance_work_mem = 2GB
checkpoint_completion_target = 0.9
wal_buffers = 16MB
default_statistics_target = 500
random_page_cost = 1.1
effective_io_concurrency = 200
work_mem = 104857kB
min_wal_size = 4GB
max_wal_size = 8GB
max_worker_processes = 8
max_parallel_workers_per_gather = 4
max_parallel_workers = 8

Problem went away. So it was a configuration issue. Probably worth updating the blog posts to cover same:
https://blog.timescale.com/analyzing-ethereum-bitcoin-and-1200-cryptocurrencies-using-postgresql-downloading-the-dataset-a1bbc2d4d992
https://blog.timescale.com/analyzing-ethereum-bitcoin-and-1200-cryptocurrencies-using-postgresql-3958b3662e51