serverfault.com

Server runs Trixie. Client runs Trixie. Local connection, my RSA key is used, passwordless login. Tor connection (Tor Hidden Service), I am asked for password. Also over Tor connection, but not local: SSH proxy fails to connect. It feels like there is a MITM attack. But could also be an SSH-over TCP issue? All of this worked under buster. Not sure about bookworm.

I'm tying to build an Ansible inventory using the Ansible amazon.aws.aws_ec2 plugin with nested hostvars. I want the resulting inventory to resemble: all: children: aws_ec2: hosts: myhost.compute.internal: .... myvargroup: mykey: 10-0-1-1 In my aws_ec2.yml file I have plugin: amazon.aws.aws_ec2 strict: True filters: tag-key: dynamic_node_role compose: foo: "'bar'" myvargroup: mykey: private_ip_address | regex_replace ('\.', '-') However when I run ansible-inventory -i aws_ec2.yaml --list -y instead of rendering the value of the expression it renders the expression itself. myvargroup: mykey: private_ip_address | regex_replace ('\\.', '-') NOTE: ansible replaced the \. with \\. in the output. The ansible documentation seems to suggest that it shou

I formatted a drive on a NAS qnap TS209 flash drive of 32 GB in ext2. Attempting to mount on debian 10 and Windows fails. Any ideas?

<INPUT TYPE=”text” NAME=”username” CLASS=”username” REQUIRED> <INPUT TYPE=”email” NAME=’emailVerify’ CLASS=’emailVerify’ REQUIRED> <INPUT TYPE=’password’ NAME=’password’ CLASS=’password’ REQUIRED> <BUTTON TYPE=’submit’ ONCLICK=”$('.cuteNess').toggle()” NAME=”submit”>Create account</BUTTON> <DIV CLASS=’cuteNess’></DIV> The terms should toggle visibility upon BUTTON click! Updated inquiry: ONCLICK=’document.getElementsByClassName(”utenhet”).innerHTML = ”Terms of Service<BR>This game (\”monkacres.se, M/A\”)<BR>Violation of the terms will produce:<BR>An error page/message”’ />Create account!</BUTTON><BR /><BR /> <DIV CLASS=’utenhet’ /></DIV> Is this the correct syntax even?

I purchased a failover /30 IP range from OVH for my Proxmox host server. Can I use all the IPs in that range and assign them to VMs on the host? ps: im using the default bridge setup Normally, with a /30 subnet, only 2 IPs are usable. However, in OVH’s setup, individual failover / additional IPs are provided as /32 and use the main host IP’s gateway. So does the same concept apply when purchasing a failover IP block (like a /30)? Can all the IPs in that range be assigned to VMs in the same way?

We run an application on Google App Engine Flexible Environment with two services: Service default: machine e2-custom-4-8704, min_num_instances: 1, disk: pd-standard Service aux: machine e2-custom-2-6656, min_num_instances: 1, disk: pd-standard Region: southamerica-east1 Runtime: custom (Docker) Network: Shared VPC between two GCP projects What happened On 2026-03-23, between 20:18 and 20:21 UTC (17:18–17:21 Brasília time), our application was completely unavailable for approximately 3 minutes. Users saw: Error: Server Error The server encountered a temporary error and could not complete your request. Please try again in 30 seconds. This was NOT a manual deploy. Cloud Audit Logs confirm the entire process was triggered automatically by GCP's own service account (GCP's internal App Engine service account) as part of the weekly OS rotation. The same behavior occurred again the

Scenario: Customer with hybrid Exchange environment; all mailboxes are in Exchange Online. Latest Exchange version (SE). Exchange was originally deployed in a resource forest topology, thus each mailbox existed in the resource forest and was linked to a user account in the main forest. Entra Connect syncs both forests and merges the Exchange attributes in the resulting user objects. All mailboxes have been migrated to Exchange Online. Now, when the customer needs to create a new mailbox, they use the following process: The mailbox is created as a linked mailbox in the on-premises Exchange server. The mailbox is them migrated to Exchange Online. From that point on, the mailbox is treated by Exchange as a remote mailbox. It would be a lot easier to simply create a remote mailbox, but it looks like this is not possible: the

If LACP declaration (server side) will still send traffic when one of the NICs fails, is there any advantage to declaring a bond as one primary one slave? Other than having the backup on a financially cheaper route I cannot see any positive to it. Is there any valid technical argument in favour? I am trying to steer our automation toward LACP as a install default rather than active_backup as it is now.

Where can I find documentation for Openvpn's new combined config/state/log file? For a number of years we have been using OpenVPN with the simple key-value pair config files with a .ovpn file extension. An added complication is that we use (short TTL) client certificates as part of the authentication. Automating the certificate refresh is trivial - we just need to update the file referenced in the configuration. However in the most recent MS-Windows version, this is replaced by a JSON file at %APPDATA%\OpenVPN Connect\config.json which appears to combine state and log data along with the CA AND client certificates. The data itself is stored within a JSON encoded string inside the JSON encoded file with added backslashes for good measure. There also seems to be some XML fragments in there too. Since the certificates and keys are embedded in this file, it apparently needs to be regenerated when the client certificate is to be updated. While it is poss

Downloading from my NAS using NFS works for low volume, but during a large file transfer the system hangs and is non-responsive to input using SSH or GUI. After transfer is complete, it will act normal. The system doesn't reboot, or crash. However, when copying files using SFTP/SMB/SCP/Rsync, the problem does not occur, only NFS. fstab: XX.XX.XX.XX:/mnt/BigMomma /mnt/BigMomma nfs auto,hard,intr,vers=4.2,rsize=4096,wsize=4096,noatime,fsc,rdirplus,tcp,actimeo=1800 Running Linux Mint 22.2 NFS to TrueNAS with Wireguard S2S tunnel. Small files work better when operating over WAN/VPN than large sizes. Optimal size is 1396 which is an ethernet packet size minus various overhead, but the IOPS increases. Anyone know the cause?