Josh M.'s questions

I have a situation where my-software requires my-common via:

Requires: my-common >= 1.2.0

During yum update my-software, my-common is not upgraded, and is left at version 0.0.1.

Why is yum not updating my-common -- and the install of my-software completes without error, even though the dependency for my-common is not resolved?

See:

# yum deplist my-software
package: my-software.noarch 1.1.0-1637255366
  dependency: /bin/sh
   provider: bash.x86_64 4.2.46-34.el7
  dependency: my-common >= 1.2.0
   provider: my-common.x86_64 1.2.0-1637243847.el7_9
   provider: my-common.noarch 0.0.1-20180703160841.el7

Note that yum thinks that my-common is resolved by either of:

• my-common.x86_64 1.2.0-1637243847.el7_9
• my-common.noarch 0.0.1-20180703160841.el7

The latter is already installed. Running yum update my-common correctly updates to the 1.2.0 version, and then my-software works correctly.

Noting that the timestamp in the latter package is incorrectly formatted and appears "newer" than the first. But the first has the version I'm requiring, so I would have expected the latter to be completely excluded from the list of valid packages.

On RHEL7, running the command (as sudo) needs-restarting hangs, as does needs-restarting -s while needs-restarting -r comes back within 2 seconds. By "hangs" I mean that it takes a really long time and eventually my SSH session times out. This was working fine until recently. What can I do to investigate the issue (barring "restart'll fix 'er")?

Relevant man page: https://man7.org/linux/man-pages/man1/needs-restarting.1.html

Note that I restarted the server and then needs-restarting completed in a couple of seconds. But this is clearly not ideal.

I'm seeing log entries in IIS's log files such as the following. Not sure where they are coming from, but the user agent is "V4+agent".

2019-08-04 22:28:45 127.0.0.1 GET /api/passphrase - 443 - 127.0.0.1 V4+agent/1.0 - 404 0 2 0 2019-08-04 22:28:45 127.0.0.1 GET /api/license/timestamp/abc123abc123abc123 - 443 - 127.0.0.1 V4+agent/1.0 - 404 0 2 15

I've followed the instructions shown at the links below, but I cannot get NodeJS v10 installed on RHEL7.

curl -sL https://rpm.nodesource.com/setup_10.x | sudo -E bash -
yum install nodejs

I am greeted with the following message:

## Installing the NodeSource Node.js 10.x repo...


## Inspecting system...

+ rpm -q --whatprovides redhat-release || rpm -q --whatprovides centos-release || rpm -q --whatprovides cloudlinux-release || rpm -q --whatprovides sl-release
+ uname -m

## Confirming "el7-x86_64" is supported...

+ curl -sLf -o /dev/null 'https://rpm.nodesource.com/pub_10.x/el/7/x86_64/nodesource-release-el7-1.noarch.rpm'

## Downloading release setup RPM...

+ mktemp
+ curl -sL -o '/tmp/tmp.p7C7NIfygt' 'https://rpm.nodesource.com/pub_10.x/el/7/x86_64/nodesource-release-el7-1.noarch.rpm'

## Installing release setup RPM...

+ rpm -i --nosignature --force '/tmp/tmp.p7C7NIfygt'
warning: /etc/yum.repos.d/nodesource-el7.repo created as /etc/yum.repos.d/nodesource-el7.repo.rpmnew

## Cleaning up...

+ rm -f '/tmp/tmp.p7C7NIfygt'

## Checking for existing installations...

+ rpm -qa 'node|npm' | grep -v nodesource

## Run `sudo yum install -y nodejs` to install Node.js 10.x and npm.
## You may also need development tools to build native addons:
     sudo yum install gcc-c++ make
## To install the Yarn package manager, run:
     curl -sL https://dl.yarnpkg.com/rpm/yarn.repo | sudo tee /etc/yum.repos.d/yarn.repo
     sudo yum install yarn

Package matching 1:nodejs-6.16.0-1.el7.x86_64 already installed. Checking for update.
Nothing to do

And yum --showduplicates list nodejs |expand shows:

Installed Packages
nodejs.x86_64                  2:9.6.0-1nodesource                   @nodesource
Available Packages
nodejs.x86_64                  1:6.12.2-1.el7                        localepel  
nodejs.x86_64                  1:6.12.3-1.el7                        localepel  
nodejs.x86_64                  1:6.14.0-1.el7                        localepel  
nodejs.x86_64                  1:6.14.2-1.el7                        localepel  
nodejs.x86_64                  1:6.14.3-1.el7                        localepel  
nodejs.x86_64                  1:6.16.0-1.el7                        localepel 

And node -v shows:

v9.6.0

Whatever I do, I can't seem to get v10 installed.

Sources:

• https://github.com/nodesource/distributions/blob/master/README.md#rpminstall
• https://tecadmin.net/install-latest-nodejs-and-npm-on-centos/

When redirecting, I want to redirect to the same scheme+host+port as the request, but with a different path. Is there an Nginx variable which is the equivalent of $scheme://$host:$port?

For example, when a request for https://www.some-site.net/login.htm comes in, it should be redirected to https://www.some-site.net/login (e.g. no ".htm").

Is there a variable that consolidates $scheme://$host:$port?

I'm using GitLab's "CI Multi-Runner", which I have installed on Windows 2012 R2 instance. Builds start correctly and begin to run, but the build is marked as complete before it even runs my entire script.

.gitlab-ci.yml

build_web:
  script:
    - cd Web
    - npm install
    - jspm install
    - gulp build

Upon pushing, the pipeline runs through npm install and then outputs "Build succeeded" and ends the build successfully (having skipped the following two commands).

I'm trying to use QEmu for Windows to convert a virtual disk image to a physical SSD. But I'm not sure about the syntax for the output_filename parameter. Here's what I tried:

qemu-img convert -p "D:\Virtual Machines\LinuxMint\LinuxMint-System.vdi" -O raw \\.\PHYSICALDRIVE5

But I get this error:

qemu-img: Could not open '-O': Could not open '-O': Invalid argument

Note that I do not have any partitions on the output drive - it's a bare drive.

Also, I only want to do this if I'll see a decent performance gain - so if it doesn't make a difference, then I won't bother.

I need to get AWS SES setup for a domain. The registrar's DNS tool only allows me to put in a block of text for the TXT/SPF records (one block per record). I'm not sure how to format the TXT record as a single block of text vs. a key/value pair.

Amazon Says

Record Type: TXT (Text)
TXT Name: _amazonses.the-domain.org
TXT Value: the-very-special-key

The registrar in this case is iPage, here's what it looks like:

So, how do I format this record as a block of text?

When AWS documentation and pricing refer to "usage" does this simply mean "if the instance is on" instead of "if the instance is on and doing work".

E.g., if I had an EC2 instance running but it was idle (CPU=0%), I assume I still get charged for that hour's usage.

In this case, if I had an EC2 instance which hosted a website (which should be accessible 24/7), it would make sense to purchase a Reserved Instance. Then, if I had to bring other instance online to share the load, those would (likely) best be served as On-Demand Instances.

Is my understanding correct?

I have a problem. A specific instance of SvcHost is using way too much ram (680MB). I've found that the following services are running under this process:

I've stopped and then started all services I was able to but the memory usage only dropped to 580MB. Here's what I found:

• Certificate Propagation - could not stop, freed 50MB while trying.
• Group Policy Client - could not stop.
• Task Scheduler - could not stop, freed 30MB while trying.
• System Event Notification Service - could not stop.
• Remote Desktop Service - didnot stop since I am remoted in.
• Windows Management Instrumentation - freed 30MB.
• Windows Update - freed 60MB.

Which of the

Is there a way to manage Windows Hyper-V Server 2008 R2 from within itself? E.g. not remotely.

I'd like to be able to manage VMs as well as connect to and work with a a VM in full screen mode.

This would provide the ability to, for instance, install Hyper-V on a laptop and then connect to the VM and work per usual - switching between running VMs in the process.

Currently I have a 2003 box running AD as the root OS on the system. I'd like to do this:

1. Backup the AD and DNS configuration on the 2003 box.
2. Wipe the drive(s) and install Hyper-V 2008 R2 as the root OS.
3. Create a new VM with 2008 R2.
4. Restore backup from step 1 onto new VM.

I can't find instructions for doing the back/restore portion. I'm hoping someone can offer some guidance.

On a Windows Server 2008 R2 Web server, I have a Windows Service which uploads files to an external FTP server. Actually multiple servers (Google, Bing, etc.). This is a .NET application which uses the built-in .NET FTP libraries.

I'm getting a mixture of the following errors:

The operation has timed out.
The remote server returned an error: (425) Can't open data connection.

Windows Firewall (Domain, Private, Public) is set to allow outbound connections.

In IIS I need to map www.example.com and example.com to use a specific SSL certificate. I have tried using appcmd and editing applicationHost.config but I don't see how to tie each host-header to a specific certificate. If I add the bindings in the config file and then look in the UI, no certificate is applied to bindings.

The certificate is for www.example.com but has an alias for example.com.