Christian's questions

We just upgraded our PostgreSQL servers to v13. We heavily use pg_dump on Ubuntu 18 systems to transfer data between databases. After upgrading the servers, pg_dump would complain about a version mismatch.

Easy enough, I installed the postgres-client-13 from the Apt Postgres repo (http://apt.postgresql.org/pub/repos/ap). pg_dump would still complain about a version mismatch so I uninstalled postgresql-client-11. After that pg_dump would complain that "PostgreSQL version 11 is not installed".

No amount or ordering of reinstalling and removing PostgreSQL clients clears up this error. Can anyone point me in the right direction to solving this issue?

In our Apache access logs, we've noticed users occasionally getting a 500 error. It is likely a database query problem within our CodeIgniter code. We can't be sure how to fix it until we see the error. So, I'm trying to get the error printed to the error log and have been failing at that. Here's where I'm at:

• Turned on error logging in Apache's php.ini file (/etc/php/7.2/apache2/php.ini) with the line: log_errors = On
• Error reporting in CodeIgniter is set to: error_reporting(E_ALL & ~E_NOTICE & ~E_STRICT & ~E_USER_NOTICE);
• The site's Apache vhosts config specifies an error log (/var/log/apache2/minans_error.log) which is getting populated with notices about IPs we've blocked in our .htaccess file, but not with 500 errors.
• The error max length in php.ini has been set to 0 with the line: log_errors_max_len = 0

• I forced log_errors on in CodeIgniter: ini_set('log_errors', 1);

• I changed the permissions on the error log file to 666.

• I forced the error log file in CodeIgniter with: ini_set('error_log', '/var/log/apache2/minans_error.log');

I'm hoping someone can tell me where I'm going wrong, suggest other things to try, or if I'm misunderstanding Apache's ability to log errors to point that out to me.

I'm setting up Nginx as a web server processing PHP via php-fpm on Ubuntu 18.04. This is working great and is lightning fast. I'm attempting to configure it to cache pages to see if performance gets even better. However, I've been unable to get it to cache pages to disk. The directory I've specified to save the pages to is currently set open to the world (777).

Here are relevant portions of my configuration files and a sample page request header:

nginx.conf:

http {
        proxy_cache_path /var/www/ma_cache levels=1:2 keys_zone=ma_cache:10m use_temp_path=off max_size=20g;
}

sites-available/mysite.conf:

server {
        proxy_buffering on;
        proxy_cache ma_cache;
        proxy_cache_valid 720h;
        proxy_ignore_headers Set-Cookie;

        location / {
                try_files $uri $uri/ /index.php;
        }

        location ~ \.php$ {
                include snippets/fastcgi-php.conf;
                fastcgi_pass unix:/var/run/php/php7.2-fpm.sock;
        }
}

Sample Response:

content-encoding: gzip
content-type: text/html; charset=UTF-8
date: Thu, 14 Nov 2019 21:08:13 GMT
server: nginx/1.14.0 (Ubuntu)
status: 200

There must be something that I'm misunderstanding or simply being stupid about. I'm hopeful someone can point me in the right direction.

I'm in the process of upgrading several of our EC2 instances from type T2 to T3. This requires enabling ENA support. I've successfully upgraded 3 of 4 instances, but the last one is having issues.

I've enabled ENA, just like the other instances, changed the instance type to T3.2xlarge, and started the instance. When I attempt to SSH into it, SSH attempts to make the connection to the instance but gets no response. I get the same result trying to make it an M5 or M4 instance as well. However, starting it as a T2 or M3, I'm able to connect to it just fine.

The OS is Ubuntu 16.04.1 LTS and ENA support is enabled:

ubuntu@ip-172-xx-xx-xxxx:/$ modinfo ena
filename:       /lib/modules/4.4.0-150-generic/kernel/drivers/net/ethernet/amazon/ena/ena.ko
version:        2.0.3K
license:        GPL
description:    Elastic Network Adapter (ENA)
author:         Amazon.com, Inc. or its affiliates
srcversion:     E19C939F9F1A3B8E900815D
alias:          pci:v00001D0Fd0000EC21sv*sd*bc*sc*i*
alias:          pci:v00001D0Fd0000EC20sv*sd*bc*sc*i*
alias:          pci:v00001D0Fd00001EC2sv*sd*bc*sc*i*
alias:          pci:v00001D0Fd00000EC2sv*sd*bc*sc*i*
depends:
retpoline:      Y
intree:         Y
vermagic:       4.4.0-150-generic SMP mod_unload modversions
parm:           debug:Debug level (0=none,...,16=all) (int)

ubuntu@ip-172-xx-xx-xxxx:/$ aws ec2 describe-instances --instance-ids i-000scrubbed000 --query "Reservations[].Instances[].EnaSupport"
[   
    true
]

Anyone have thoughts/ideas?