Well, after looking for many solutions. I came here now.
I am setting up WHM/cPanel for hosting website. Everything was going smooth but I am stuck on FTP connection (Server sent passive reply with unroutable address. Using server address instead.)
Server Details:
CentOS Linux release 7.2.1511 (Core)
WHM/cPanel Version 11.58.0.13
FTP Server: PureFTPD
Acutal error while connecting
To fix this issue and get FTP working you need to open up more numbered ports so FTP can connect. I assume you are using CSF.
Login to WHM then go to CSF >> Firewall Configuration >>
allow TCP_In 30000:50000 and TCP_Out 30000:50000
Once you made the changes Restart the firewall
Now you need to make changes in FTP config file to use these ports, you will find this file to this location /etc/pure-ftpd.conf
Now you will see a line as follows and you will need to uncomment it
# Port range for passive connections replies. - for firewalling.
PassivePortRange 30000 50000
Restart FTP Service and should work.
Part of an install includes a postgres database install. The database is running and the App that uses it is connecting fine.
I want to use the postgres install for some of my own reporting needs. I've got a fair way to getting to the database but I am stumped at the last bit.
I can do a psql -h /path/to/socket but get
psql: FATAL: authentication failed for user "postgres": invalid authentication method.
I have gone in to the pg_hba.conf and the only lines not commented are:
local all all trust
host all all 127.0.0.1/32 trust
host all all ::1/128 trust
As far as I can see this should mean any local attempt at connection would be trusted and therefore OK.
I've tried with specifying different users (root, me, etc) and always get the same reply. Any ideas how I can access the server?
I just installed postgresql-9.1.4-1-windows-x64 on a Windows 7 64 bit machine. I'm having trouble starting the service and connecting to a database.
After a successful installation I've tried the following based on similar postings.
1) Looked for "Start Server" under Start > All Programs > PostreSQL 9.1 and could not find it.
2) Tried starting the server from the command line
pg_ctl.exe -D "C:\Program Files\PostgreSQL\9.1\bin\data"
This gave me the error:
could not create lock file 'postmaster.pid': Permission denied
I have administrative rights, and there is not postmaster.pid file either in the bin or data directories.
3) Next I tried starting the Service from Admistrative Tools by right clicking on the postgresql-9.1.4-1-windows-x64 Service and selecting Start. I received the message:
The postgresql-9.1.4-1-windows-x64 Service on local computer started
and stopped. Some services stop automatically if they are not in use
by other services or programs.
The Event Viewer showed the error as Timed out waiting for server startup
4) I figured there the data in the data directory was probably and initial database, but just in case I ran "initdb" and got:
If you want to create a new database system either remove or empty the
directory c:\program files\postgreSql/9.1/data or run initdb with an
argument other than c:\program files\postgreSql/9.1/data
4) And just for fun I Started pgAdminIII, right clicked on "PostreSQL 9.1(localhost:5432)", selected Connect, entered password, and got:
could not connect to server: Connection refused (0x0000274D/10061) Is
the server running on host "localhost" (::1) and accepting TCP/IP
connections on port 5432?
Does anybody have a suggestion?
Thanks.
«"could not create lock file 'postmaster.pid': Permission denied"»
Do not look any further, Postgres cannot start if it can't create this temp file. If it is not created, you evidently cannot find it if you look for it on the disk. Your DATA directory has been created so no need to re run initdb again and if you try to use pgadmin it complains that it cannot connect to Pg -- which is not running.
I am not familiar with windows but found out where postmaster.pid is to be created you will probably find out why Postgres cannot create this file.
Hope it helps.
I have no trouble running dbgsrv on one machine, and then from another machine using cbb to attach to the remote session with
cdb -premote tcp:server=ipaddress,port=port process
However for some reason I can't make
cdb -QR ipaddress
work to show the server. Even with the server running on the same machine, firewall off, running
cdb -QR 127.0.0.1
returns
Unable to query 127.0.0.1
Similarly windbg > Connect To Remote Stub > Browse also fails to return anything.
Also, while I can connect to the remote session using
windbg -premote tcp:server=ipaddress,port=port process
I can't use windbgs "Connect to remote stub" to connect to the very same session using connection string
tcp:server=ipaddress,port=port process
Any idea what is causing me to not be able to query for server sessions? Also why can I connect from the command line, but not from within windbg using Connect To Remote Stub?
P.S. This should probably also be tagged dbgsrv, but there is no such tag.
P.P.S. The reason I want to use dbgsrv rather than .server command is that I want to resolve symbols and source on my local machine.
The correct syntax is cdb -QR \\127.0.0.1
When using windbg > Connect To Remote Stub > Browse, Enter the IP address of the computer you want to browse in the Machine field before clicking on Refresh
Hello I've got a problem accessing Oracle DB from our datacenter through a tunnel.
We've got a pretty standard datacenter with one machine being accessible from the outside
(I put it's IP in the /etc/hosts file as dc) and the Oracle DB inside. The IP address of our oracle database on internal network is 192.168.1.7
To create a tunnel I'm using the command:
ssh -L 1521:192.168.1.7:1521 root#dc
and of course it works (sometimes I also add some debug -vv to see if anything is passing through).
Now the difficult part - connecting to Oracle. I installed instantclient 11.2. and my tnsnames.ora looks like that:
testdb =
(DESCRIPTION =
(ADDRESS_LIST =
(ADDRESS = (PROTOCOL = TCP)(HOST = localhost)(PORT = 1521))
)
(CONNECT_DATA =
(SERVICE_NAME = dbname)
)
)
And when I try to connect using the command:
./sqlplus username/pass#testdb
It starts connecting through the tunnel (I see it in the ssh debug) but then it fails
telling:
./sqlplus username/pass#testdb
SQL*Plus: Release 11.2.0.1.0 Production on Wed Jan 13 20:46:07 2010
Copyright (c) 1982, 2009, Oracle. All rights reserved.
ERROR:
ORA-12541: TNS:no listener
Enter user-name:
When I'm trying to execute this same command on when I'm on the intranet it works (obviously the only difference is that in the tnsnames.ora HOST we have 192.168.1.7 and not the localhost).
I also tried to use the simple command line:
./sqlplus username/pass#//localhost:1521/testdb
or alternatively
./sqlplus username/pass#//localhost:1521/testdb
But nothing helped :)
I would appreciate any help or suggestions. Am I missing some ssh flag to make it possible?
Probably the log file:
***********************************************************************
Fatal NI connect error 12541, connecting to:
(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=TCP)(HOST=localhost)(PORT=1521)))(CONNECT_DATA=(SERVICE_NAME=DBNAME)(CID=(PROGRAM=sqlplus#velvet)(HOST=velvet)(USER=johndoe))))
VERSION INFORMATION:
TNS for Linux: Version 11.2.0.1.0 - Production
TCP/IP NT Protocol Adapter for Linux: Version 11.2.0.1.0 - Production
Time: 13-JAN-2010 20:48:42
Tracing not turned on.
Tns error struct:
ns main err code: 12541
TNS-12541: Message 12541 not found; No message file for product=network, facility=TNS
ns secondary err code: 12560
nt main err code: 511
TNS-00511: Message 511 not found; No message file for product=network, facility=TNS
nt secondary err code: 111
nt OS err code: 0
where velvet is my local hostname and johndoe is my local username.
Why is it sent to the other side?
UPDATE:
After investigating a little bit more from inside datacenter and it looks like:
- the first connection is going to the port 1521
- but then sqlplus is redirected to the port number > 3300, which is different everytime and incrementing by 3 (at least few tries I had)
- when we are trying to connect trough a tunnel sqlplus will try to connect to localhost and it will obviously fail
So the error "No Listener" comes probably from the fact that we are not redirecting those ports. Is there any way (probably some option in tnsnames.ora file) to force some specific port to be used?
Look into Metalink ID 361284.1 (Edit: effectively not public, but find the info here)
It seems like Oracle Connection Manager would be your option. It basically handles the port redirects inside the firewall. I haven't used it before, so cannot advise you further.
Update: Another way to go would be to use MTS, configure dispatchers with certain ports and open these ports in the firewall. You wouldn't have to install additional software for this, but connecting through shared server may require increasing LARGE_POOL_SIZE, among other considerations. So you'd still need the DBA role to change the DISPATCHERS parameter. You'd also have to bounce the DB.
Normally this should work. I would not use a default listener port as an entry for the ssh tunnel but that should not be the problem. I would also not user the root account to create the ssh connection, preferably a dedicated regular account. Are you using shared servers or does the database happen to be a RAC database with a load balance configuration?
A nice explanation is here How can I connect to ORACLE DB through ssh tunnel chain (double tunnel, server in company network) ?, a bit more complicated .....
update
checkout DbVisualizer, it now has integrated ssh tunneling. I think it is worth to al least give it a try. It's not free but good. Multi platform and multi database and very flexible.
In my case the problem is that the DB server has several IPs and when I used SSH tunnel it was connecting to wrong different one.
So try to check, if the destination IP is the same as the IP in the listener.ora file on the DB server.
Can you try to make a trace to determine exactly what is happening:
For server trace, try here (be carefull! all the new request will be traced and the server can be collapsed).
For client trace, checkout here.
MJ! Your tunnel is only for the initial tcp connect, your own LISTEN port is not tunnelled, and probably unimplemented. Firewall should allow a connect back to you, similar to active FTP.
All ports for Oracle are documented quite extensively starting page 670 of "Building Internet Firewalls" 2/E Chapter 23, paragraph: Oracle SQL*Net and Net8. You can view it on SafariBooksOnline.com
ISBN 1565928718
Perhaps your listener haven't been started yet. Try run "lsnrctrl start" command.
Also a good explanation is here connection to an oracle database though a SSH secure shell which worked for me.
Open putty and on the session page, enter the name of a server and make sure SSH is checked. The server can be any server that you have a
username and password to login with. I use one here called BLUEBIRD as
I own it!
On the connection->ssh->tunnels page, uncheck both options at the top ("Local ports accept ..." and "Remote ports do the same").
Enter 9999 (or any port above 1024 as the Source Port.
In the destination, enter the database host and port as per tnsnames. In my case, this is a server called GREENBIRD and a port of
Enter this as server:port.
As the port being forwarded is on your desktop, check the "Local" option. Leave "Auto" checked as well for the IP version.
Click the Add button. You will see L9999 greenbird:1521 (your will differ) in the list of forwarded ports.
Go to the session page again, Enter a name for your saved session and click save.
Click open. Supply a username and password for the server (BLUEBIRD in my case). You will login a normal ssh session to the server named
BLUEBIRD.