DA not working after upgrading to 1.293

[asekeris]

Update worked but messed up the advanced skin.
Only logo is diplayed now.
In de standard user layout the advanced button is broken.
No solution yet need help.

[SametYilmaz]

Edit: I found the bug...
do this:
Login trough SSH:
nano /usr/local/directadmin/data/skins/enhanced/header.html
(or your editor)
Now search for the line

<td valign=top>
<table width=100&#37; height=100% cellpadding=0 cellspacing$
<tr>
<td height=120 align=left background="/$
</tr>
<t

As you can see it says <t and not </td>
Change that <t tot </td> and its fixed :)

[SuperHost3000]

Well, there are more problems with the enhanced skin.. I hope that the problems soon are fixed, my clients are not happy...

[asekeris]

Edit: I found the bug...
do this:
Login trough SSH:
nano /usr/local/directadmin/data/skins/enhanced/header.html
(or your editor)
Now search for the line

<td valign=top>
<table width=100&#37; height=100% cellpadding=0 cellspacing$
<tr>
<td height=120 align=left background="/$
</tr>
<t

As you can see it says <t and not </td>
Change that <t tot </td> and its fixed :)

Removing the '<t' complete works better.
Thanks for the tip admin and reseller repaired partial.
There are some other issues with broken links I think DA has to fix.

[elkabong]

After upgrading from 1.292 to 1.293 we have noticed that the information on the right side under User Level Access is gone! There are no links or information there only under Admin and Reseller Access.

This bring me to this question - Is DA 1.293 broken? Has anyone got a similar issue?

regards,

Errol

[pongzaa]

Yes, after i upgraded to 1.29.3
My control panel interface is error.
no menu anymore... DA pls. correct it as soon as possible.

[nobaloney]

We've verified that there appears to be an issue with the update. Please continue to discuss this issue here. DirectAdmin staff has been notified of the problem and of this thread.

Jeff

[tristan]

What exactly isn't working? I just tested this on a Debian box with 1.293 but can't find any problems in the user level.

[nobaloney]

I recommend that if you haven't already joined the da-announce mailing list to be immediately notified of updates and issues, this may be a good time for you to do so.

The da-announce list, which we host, is a standard majordomo list.

To join, send an email to: majordomo@daserversolutions.com.

Majordomo lists don't care about the subject, so we don't care what you use, but the server will send you back a reply, so you should use whatever subject will help you find the reply.

The body of the email should be:

subscribe da-announce address@example.com
Of course replace address@example.com with your email address.

If you don't get anything back from the server then whitelist email from this address: da-announce@daserversolutions.com.

Jeff

[nobaloney]

I'm glad to hear it's working for you, tristan.

The problem we've seen is exactly as described by elkabong.

On CentOS 4.4.

I'll be very happy if this is a false alarm.

I'll test further later today; it's 10:24 am Sunday morning here.

Jeff

[Icheb]

Thank you... I thought it was just me...

On CentOS 4.4, fresh install.

User level main page seems broken.

[ZPillay]

Problems here too with the template.
CentOS 4.4

After the upgrade the template gets screwed: Header repeated on main page untill it reaches the footer (so no content shown, just a repeated header and the footer).

Note: Problem only occurs in Internet Explorer. No problems in FireFox.

[kevinb]

This appears to be working fine for me on Centos 4.4 in both Firefox and IE7. Are the people with issues using IE6?

[elkabong]

The DA control panel homepage is screwed up in IE7 (The DA header is broken up into 4 pieces and covers the screen and there is no access to anything). At least the control panel is "readable" in Firefox but with no information in the User Level Access Panel!

My OS is also CentOS 4.4-32

Errol

[asekeris]

Problems are the same as other repliers stated also on Centos 4.4-32
The change in the html file gives a quick and dirty repair so at lest admins and resellers can do most of their things. :D

[nobaloney]

Problems in any other versions besides CentOS?

Problems with any other skins besides Enhanced?

Note we also noticed a problem where restoring backups manages to break the /etc/httpd/conf/ips.conf file, which means httpd won't start.

I bring that up because both problems appear similarly; they're both caused by text which appears out of place, in a file.

Jeff

[asekeris]

Also noticed a damaged button in the default skin.

Butto 'advanced'

[DirectAdmin Support]

Thanks for the report.
I checked our centos 4.4 box and everything is running correctly over the network file system.

I extracted the es4.4 update.tar.gz and can confirm that it did not compress correctly. Why exactly, I can't say for sure, but I've recompressed it, double checked it's contents and it's appears to be fine now. The package is now updated, so for anyone affected (only looks like the 4.4 package), you can run this guide to get the new update.tar.gz:

http://help.directadmin.com/item.php?id=29

Note that tar did not return any errors when the bad package was created, so the problem was likely the network file system not giving all the data it should have. I'll keep an eye out for anything else.

Note, some people who have a 4.4 system are actually using the 4.0 package which is fine, hence you're not affected.

John

[nobaloney]

Thanks for your prompt response, John.

Everyone else, post your success stories :) .

Jeff

[elkabong]

I did the update as in the link and everything went well until I got to:

[root@www1 scripts]# service directadmin restart
bash: service: command not found

should be: /etc/rc.d/init.d/directadmin restart

I logged back into DA (via both Firefox and IE7) and I can confirm that my User Level Access is back! However, httpd service still remains broken and if I try to restart it from the control panel I am getting this error:

An error has occurred

Details

/sbin/service httpd start 2>&1

If I try restarting from the command line:

[root@www1 admin]# /etc/rc.d/init.d/httpd restart
Stopping httpd: [FAILED]
Starting httpd: Syntax error on line 30 of /etc/httpd/conf/ips.conf:
Invalid command '|*if', perhaps mis-spelled or defined by a module not included in the server configuration
[FAILED]
[root@www1 admin]#

Any ideas?

[smtalk]

Try:


/etc/rc.d/init.d/directadmin restart

[elkabong]

smtalk,

Thanks for that.

[DirectAdmin Support]

If the httpd.conf files need rewriting, type:
echo "action=rewrite&value=httpd" >> /usr/local/directadmin/data/task.queue
echo "action=rewrite&value=ips" >> /usr/local/directadmin/data/task.queue
/usr/local/directadmin/dataskq dJohn

[elkabong]

Thanks again John httpd works fine now.

[asekeris]

:) Thanks for the quick response.

Everything works fine now.

[hostpc.com]

We just encountered our first real problem with this upgrade

The file /etc/httpd/conf/ips.conf was completely munged with one users partial httpd.conf file appended to it. Not sure HOW that could have happened, but it took apache right offline (obviously).

A simple edit repaired the file (just deleted the erroneous lines), but it definately caught our attention.

[jissh]

Are the issues with this update restricted to CentOS 4 or other OS's as well? Just want to know what OS's are effected before we update, a list off all effected OS's would be great.

Thanks

[hostpc.com]

cat /etc/redhat-release
CentOS release 4.4 (Final)

[DirectAdmin Support]

There was 1 corrupted update.tar.gz file.
If your *license* is set to centOS 4.4, and you updated within 24 hours of the release, then you'd need to get a new update.tar.gz (see page 1). Not all 4.4 machines use 4.4. license. Some have the 4.1 license which worked fine. Either way, all OS's will be fine now as it was quickly fixed once identified.

Also, if apache went down, then make sure you have the fixed update.tar.gz, then run the echo commands (page 1 again) to rewrite the ips.conf and httpd.conf files. You'll know if your update.tar.gz is bad by logging into DA and looking at the enhanced skin. You'll have the header repeated many times vertically.

The issue was caused by a network file system error that didn't show up as an error when compressing the the package, else the packaging system would have complained.

John

[hostpc.com]

Thanks John... that seems to have fixed the issue.

Ironically, there's several machines with the exact same config (updated at the same time) - only one was affected. I guess I owe some uptime credits to customers.

Thanks for the quick fix

[NVarra]

Obviously issues like this are never good, but I still stand that updating/upgrading/managing DirectAdmin boxes is by far the least stressful part of my job.

We finally ditched our last Plesk box a few months ago and I've never been happier. I look forward to new DirectAdmin releases, where I dread seeing new Plesk and cPanel releases.

Good work guys.

[Phantium]

Hmm, i'm still afraid to update. :(

[smtalk]

Phantium, there is no reason for that :) The archive has been fixed and there are no bugs now.

[Phantium]

Phantium, there is no reason for that :) The archive has been fixed and there are no bugs now.

Okay thanks SMTalk. :)

[DirectAdmin Support]

For anyone who is paranoid, you can backup your update.tar.gz before updating:
cd /usr/local/directadmin
cp update.tar.gz update.tar.gz.backupThen use the manual update procedure. If there are any problems, just extract the update.tar.gz.backup, and run "./directadmin p", then restart DA and you're back to normal.

But... as smtalk mentioned, the issue is gone now ;)

John

[agnivo007]

If I try to update from DA Licensing/Updates section, everytime I see that 1.29.2 is staying instead of being upgraded to 1.29.3

Also, I'm getting this error while trying to do manual update.

[root@helios directadmin]# wget -O update.tar.gz https://www.directadmin.com/cgi-bin/daupdate?uid=[nnnn]\&lid=[nnnnn]
--08:53:07-- https://www.directadmin.com/cgi-bin/daupdate?uid=[nnnn]&lid=[nnnnn]
=> `update.tar.gz'
Resolving www.directadmin.com... 66.51.122.131
Connecting to www.directadmin.com|66.51.122.131|:443... connected.
Segmentation fault

[DirectAdmin Support]

what OS are you trying to update?

For wget, try recompiling:
http://help.directadmin.com/item.php?id=119

John

[pucky]

The moral of the story. Never have auto updates checked in DA either.

[agnivo007]

what OS are you trying to update?

For wget, try recompiling:
http://help.directadmin.com/item.php?id=119

John

CentOS 4.4 32bit

Tried to recompile wget 1.10.2 as suggested, but fails. A snippet of errors is as below :


make[1]: Entering directory `/wget-1.10.2/po'
file=./`echo bg | sed 's,.*/,,'`.gmo \
&& rm -f $file && PATH=../src:$PATH msgfmt -o $file bg.po
/bin/sh: msgfmt: command not found
make[1]: *** [bg.gmo] Error 127
make[1]: Leaving directory `/wget-1.10.2/po'
make: *** [install.mo] Error 2


After a little research, I found that msgfmt is a part of gettext package...but need help to solve this.

BTW, recently I had updated OpenSSL to 0.9.8e

[smtalk]

Try
# rpm -e wget
# yum install -y wget

[agnivo007]

Great...worked clike a charm...feels great when you learn things with the experts!


Installed: wget.i386 0:1.10.2-0.40E
Complete!
[root@helios /]# wget --version
GNU Wget 1.10.2 (Red Hat modified)

But still :

Connecting to www.directadmin.com|66.51.122.131|:443... connected.
Segmentation fault
And update through DA panel fails.

If I try notmal http url to get the update, the file content is :

[root@helios directadmin]# cat update.tar.gz
* You are not allowed to run this program *