Felix's questions

I am trying to install Microsoft SQL Server on Ubuntu 21.10 and I am getting an error. I noticed that Microsoft page doesn't mention 21; so maybe it's not possible yet. But Canonical announcement specifically mentions SQL Server support.

Anyway, when I run sudo apt install mssql-server I am getting the following:

The following packages have unmet dependencies: mssql-server : Depends: libldap-2.4-2 but it is not installable

if I try to install libldap manually, it just says that unable to locate package. However, when I run sudo apt install libldap-2.4-2, I get something more interesting:

Package libldap-2.4-2 is not available, but is referred to by another package.
This may mean that the package is missing, has been obsoleted, or
is only available from another source
However the following packages replace it:
  libldap-common

But how do I tell sql server installer to use it?

Maybe I am using a wrong repository? This guide still uses 20.04 - and there isn't much relevant under 21.

I have a new Centos 8 box. I also have Windows 2016-based domain controller that serves all my windows servers and workstations (about 10 altogether). I successfully joined AD - or at least it looks like it was successful. Cockpit shows correct domain, and I can do kinit:

# kinit adadmin
Password for [email protected]
#

(and if the password is wrong, I get kinit: Password incorrect while getting initial credentials as expected).

However, I cannot connect with as AD user. I also get the following errors:

# id adadmin
id: `adadmin': no such user

and

# net ads info
ads_connect: No logon servers are currently available to service the logon request.
Didn't find the ldap server!

Finally, I get the following error in /var/log/messages:

Nov 11 22:32:43 centos8 sssd[be[example.com]][12694]: GSSAPI Error: Unspecified GSS failure.  Minor code may provide more information (Server not found in Kerberos database)

I tried different things in krb5.conf and sssd.conf - but it didn't make a difference

I have IIS 10 running on Windows 2016 (I see exactly same behavior on Windows 2019). I enabled static compression. I can see that the browser (Chrome) sends accept-encoding: gzip, deflate, br. I can even see that a compressed version is generated in C:\inetpub\temp\IIS Temporary Compressed Files\JMSv2\$^_gzip_D^\WEBSITES\JMSV2\MAIN.3616F3D5444DF5AB2F98.JS (5MB in size). Everything seems to be working properly. However, javascript file comes uncompressed -

content-length: 38886877  
content-type: application/javascript

What am I missing? I saw in one answer that corporate network was stripping headers from response. My server is on AWS and standalone. Dynamic caching is not enabled

I have MS SQL Server 2017 running on CentOS for several months without a problem. Three days ago I applied yum update that included update for mssql-server as well. and ever since I can't connect to the server, getting the following error:

Sqlcmd: Error: Microsoft ODBC Driver 17 for SQL Server : Login failed for user 'sa'. Reason: Server is in script upgrade mode. Only administrator can connect at this time..

Most of the recommendations are to wait until script updates finish; but I waited for 3 days. Also restarted the service and restarted the box.

Ever since CentOS 7 came out joining machine to AD has been a breeze. It just works, and if it doesn't, the reason is obvious from the error message. However, now I am stuck:

# realm join [email protected] example.net -v
 * Resolving: _ldap._tcp.example.net
 * Performing LDAP DSE lookup on: 192.168.1.50
 * Successfully discovered: example.net
Password for [email protected]:
 * Required files: /usr/sbin/oddjobd, /usr/libexec/oddjob/mkhomedir, /usr/sbin/sssd, /usr/bin/net
 * LANG=C LOGNAME=root /usr/bin/net -s /var/cache/realmd/realmd-smb-conf.R9517Y -U [email protected] ads join example.net
Enter [email protected]'s password:
Failed to join domain: failed to join domain 'example.net' over rpc: The directory service was unable to allocate a relative identifier.
 ! Joining the domain example.net failed
realm: Couldn't join realm: Joining the domain example.net failed

It seems that the culprit is this directory service was unable to allocate a relative identifier error; but anything I can find relates to connecting to DC after joining.

Domain Controller is Windows 2016 on 192.168.1.50

Interesting (and not surprising, give that domain was successfully discovered), realm discover is successful:

# realm discover example.net
example.net
type: kerberos
realm-name: EXAMPLE.NET
domain-name: example.net
configured: no
server-software: active-directory
client-software: sssd
required-package: oddjob
required-package: oddjob-mkhomedir
required-package: sssd
required-package: adcli
required-package: samba-common-tools

Also, if I use wrong password or use another account that doesn't have join privileges - I get the error that I expect.

When I run aws command help it pipes the output through the pager. It has it's pros and cons; the biggest con is that once I press Q to exit, everything is gone from the screen (yes, I know, I can crash it by Ctrl-C).

Is there a way to just output it to the screen the way it works in Windows command prompt? With man I can run it as man -P cat mv, but I couldn't find anything like that in aws help

I am running Grid Scheduler (fka Sun Grid Engine) on Amazon Web Services. Master node is running all the time, but I want to programmatically add nodes to the cluster (also remove - but remove is not a problem).

I launch an instance from existing AMI; during startup it POSTs it's hostname (say ip-10-11-12-13) to the master node. So, if I were doing it manually, I'd run qconf -ae and fill the template with the hostname. I would also run qconf -mconf all.q and add the new host and its number of core to the SLOTS statement.

I guess I can build a file and use qconf -Ae fname and qconf -Mconf file_list. Is there a way to do such updates without awk/sed file tinkering?

Edit: I am familiar with MIT StarCluster; but for variety of reasons we are not using it.

I am running a Centos machine in Amazon cloud. Suddenly, I can't ssh into it. Fortunately, there is R Studio running that includes an ability to run BASH shell. So, I see in /var/log/boot.log that sshd failed to start.

When I run it from the command line sudo service sshd start I get an error that Starting sshd: /etc/ssh/sshd_config: Permission denied. I tried to set sshd_config permissions to either 644 or 600 - but I get the same error. Also I tried sudo su - and then start service.

And it is not the limitation of the shell itself: I can start httpd without any problems.

I don't even know what else to try...

I have a small office network with router (running OpenWRT), Windows Domain Controller (used to be 2008R2; I just backed it up and upgraded to 2012), about a dozen AD clients (3 server and windows workstation) and several non-AD clients (network printer, PBX).

The problem is that the clients can't access servers by name (only by IP). I tried all kind of permutations. Right now domain controller runs DNS server for all desktops; but unless I put an entry in hosts file - I can only get by IP. I have router as DHCP server (since not all devices are on AD); and except for Domain Controller all IP addresses, including "static", are assigned by the router.

Most frustrating, some servers sometimes just work! for example, I can often get to the Linux box by name (it is part of Domain using Beyond Trust Integration Services); but I can never get to SQL Server box. Seems like non-domain devices see more names than domain members...

This network should be fairly typical; but I couldn't get any guidance about how to set up DNS/DHCP service to make all nodes happy. The closest is this question, but still it's different!

Thanks

I am trying to connect from desktop to SQL Server 2012 (on a different server) and getting the following error:

Connection Timeout Expired. The timeout period elapsed during the post-login phase. The connection could have timed out while waiting for server to complete the login process and respond; Or it could have timed out while attempting to create multiple active connections. The duration spent while attempting to connect to this server was - [Pre-Login] initialization=190; handshake=4; [Login] initialization=0; authentication=102; [Post-Login] complete=14002; (Microsoft SQL Server)

The client is SQL Server Management Studio 2012. Firewall on the server is turned off. I can telnet to port 1433. And connecting from Studio on the server itself works fine. Desktop and Server are on the same segment; using Windows authentication.

Both Windows Server 2008R2 and SQL Server are just built from scratch.

What puzzles me is that post-login phase. I have a feeling that I am looking in the wrong place (that is, network connectivity); but don't know what the right place would be...

I tried to install SQL Server 2012 Standard on Windows 2008 and on Windows 7 (I only wanted to install management studio on Win7). Same ISO; both machines are newly built. However, on Windows 7 Shared features directories are set to C:\Program Files... and greyed out even though none of the options are checked. C: is OS drive; however, it's a rather small SSD, so I am trying to install everything I can to another drive.

On Win2008 it is not greyed out (unfortunately, I don't need to change anything there).

What could cause Shared directories to be greyed out on Windows 7?

I installed XenServer 6.0 on a machine with 2 drives (80GB SSD and 1TB SATA). My goal is to have some space for local storage where I can put ISO (since it will be very difficult to access the box physically, and I would prefer to install VMs from ISOs and not from CDs).

However, no matter what I specify for LVM, the only filesystem on sda is 4G (and only 2GB is available). XenCenter shows me plenty of disk space available for VMs; but that doesn't help me with ISOs.

Is it possible to increase local storage? I haven't started building VMs yet - so now is ideal time to partition the disk the way it should be. But parted is not installed - and I suspect for a reason.

Here is the output from pvdisplay:

[root@xenserver ~]# pvdisplay
--- Physical volume ---
PV Name               /dev/sdb
VG Name               VG_XenStorage-a7bbe7ea-f205-5f25-4ef8-daec4e08bcb2
PV Size               931.51 GB / not usable 6.41 MB
Allocatable           yes
PE Size (KByte)       4096
Total PE              238464
Free PE               238463
Allocated PE          1
PV UUID               yqrKQi-l8mD-k3mo-PxEs-M0bK-7xx3-17nnzE

"/dev/sda3" is a new physical volume of "66.53 GB"
--- NEW Physical volume ---
PV Name               /dev/sda3
VG Name
PV Size               66.53 GB
Allocatable           NO
PE Size (KByte)       0
Total PE              0
Free PE               0
Allocated PE          0
PV UUID               HlXvAo-f6gw-eqMz-5Y1d-YzYi-19tK-xZTWoE

So, the real question is, how can I make use of 66GB on /dev/sda3 without breaking anything in XenServer?

I have an application running on Tomcat 6, that (from Tomcat perspective) doesn't have any authentication. Also, it needs to run over SSL. Last, one of the pages allows you to save an Excel file. Saving file fails with an error described here: Pragma No-Cache prevents the file from being saved, and then browser complains about connection lost. This happens on IE 7 and 8; doesn't happen in Firefox or IE 9 (unfortunately most of the users have IE7).

Tomcat has attributes disableProxyCaching and securePagesWithPragma that work around caching issues in some browsers however, these are attributes for various authenticator valves.

My question is, how can I prevent Tomcat from adding a Pragma: no-cache header when there is SSL, but no authentication?

EDIT: I tried nonloginauthenticator valve - but it doesn't seem to make any difference, and I don't see it in Tomcat 6 configuration administration anyway (although I see it in APIs)