We need to buy some memory for some servers. I have three quotes in total. One for memory from dell, one for Kingston memory and one for Crucial/Micron memory.

The one for crucial memory and Kingston memory are half the price as the same memory from Dell.

I inquired with Dell why their memory is 2x more expensive than alternatives and the answer I got back was that:

"Dell memory is more expensive as it is tier-1 OEM grade, which is a much higher quality grade of RAM. The warranty for Dell memory also goes through Dell tech support."

The warranty part totally makes sense to me - I get that purchasing through Dell means I get 4 hour turnaround time on replacement memory.

However, what is "Tier-1 OEM grade memory"? Is there such a thing? And if there is - is it a "much higher quality" of memory? Or is this all a lot of malarkey to justify charging twice as much?

I can't find much about tier one OEM grade memory so I'm inclined to think this is nonsensical double speak but I'm wondering if they know something I don't.

I have the following in my apache config file:

Define THUMBS_ROOT "//mysmbserver/website/thumbs"
Alias "/thumbs" "//mysmbserver/website/thumbs"
<Directory {THUMBS_ROOT}>
    Options Indexes FollowSymLinks
    <RequireAny>
        Require ip 66.54.56.34
        Require ip 159.101.84.4
    </RequireAny>
</Directory>
LoadModule authz_core_module modules/mod_authz_core.so

My apache version is Apache/2.4.29 (Win64). (Yes I realise this is an old version of Apache - I can't upgrade due to it being an appliance device)

Even those I've added RequireAny and RequireIP apache is ignoring this and allowing any IP to access the thumbs directory. I can't figure out why. Can anyone help me understand?

Further complicating matters I have a load balancer device sitting in front of apache so really what I need to be looking at is X-Forwarded-For not the source IP. I'm not sure offhand if I need to do anything for RequireIP to detect X-Forwarded-For?

Either way though its allowing access regardless of IP.

Edit: I've also tried:

LoadModule authz_core_module modules/mod_authz_core.so
Define THUMBS_ROOT "//mysmbserver/website/thumbs"
Alias "/thumbs" "//mysmbserver/website/thumbs"

SetEnvIF X-Forwarded-For "66.54.56.34" AllowIP1
SetEnvIF X-Forwarded-For "159.101.84.4" AllowIP2

<Directory {THUMBS_ROOT}>
    Options Indexes FollowSymLinks
    <RequireAny>
        Require env AllowIP1
        Require env AllowIP2
    </RequireAny>
</Directory>

That doesn't work either.

I am trying to record the a client ip stored in the X-Forwarded-For http header in the Apache access logs but having no luck and have been pulling my hair out for hours. I've tried a ton of different things - nothing seems to work!

First I tried:

# Fields to record
LogFormat "(%{X-Forwarded-For}i) %l %u %t \"%r\" %>s %b \"%{Referer}i\" \"%{User-Agent}i\"" combined

# Log locations
CustomLog "/var/log/httpd/access.log" combined env=!dontlog

I also tried

# Fields to record
LogFormat "%a %l %u %t \"%r\" %>s %b \"%{Referer}i\" \"%{User-Agent}i\"" combined

# Log locations
CustomLog "/var/log/httpd/access.log" combined env=!dontlog

I also tried:

RemoteIPHeader X-Forwarded-For
RemoteIPTrustedProxy 192.168.1.10


# Fields to record
LogFormat "%a %l %u %t \"%r\" %>s %b \"%{Referer}i\" \"%{User-Agent}i\"" combined

# Log locations
CustomLog "/var/log/httpd/access.log" combined env=!dontlog

This is on CentOS8

httpd -v
Server version: Apache/2.4.37 (centos)
Server built:   Sep 15 2020 15:41:16

Remote IP module is installed and I assume enabled?

 httpd -M | grep remoteip
 remoteip_module (shared)

I know the X-Forwarded-For header is being sent because I did a tcpdump:

    User-Agent: Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:82.0) Gecko/20100101 Firefox/82.0
    Accept: text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8
    Accept-Language: en-CA,en-US;q=0.7,en;q=0.3
    Connection: keep-alive
    Cookie: _fbp=fb.1.1599523605116.597747458; __qca=P0-681804816-1599523605211; _ga=GA1.2.489007183.1600273387; _gid=GA1.2.315615129.1604091772; LB=1460283402.20480.0000; _dc_gtm_UA-34638206-1=1; _gat_UA-34638206-1=1
    Upgrade-Insecure-Requests: 1
    Pragma: no-cache
    Cache-Control: no-cache
    X-Forwarded-For: 69.165.232.76
    X-Forwarded-Proto: http

Case/spelling is identical to what I've put in my conf. So why do I keep getting the load balancers IP instead of the actual client IP?

I know I'm editing the right file because if I break the file by putting gibberish in it apache won't load.

I'm at my wits end. Someone please help :)

I used the VMware standalone converter 6.2 to convert a windows 2012 R2 system to a VMware ESXi VM.

The conversion reports 100% success and I can see the VM in vCenter. When I start it up though I just get a black screen with a white swirly at the bottom. I've left the system for hours on end and it never starts up.

I detached the VMDK file and attached it to another working VM and mounted the drive. I can see all the files and both partitions. So I know the clone did successfully clone my files.

I next rebooted the system with the windows 2012 R2 installation media and then went to the recovery console where I ran the following commands:

Bootrec /Fixmbr 
Bootrec /Fixboot
Bootrec /rebuildBCD

Then restarted my system - same result.

I've tried conversions that don't install vmware tools, that do install vmware tools, that don't reconfigure the destination vm, that do reconfigure the destination VM.

I feel like there must be something minor preventing a boot but I'm at a loss as to what else to try. Does anyone else have any suggestions?

Thanks Brad

I have the following entry in my apache config file:

<LocationMatch "/hls-vod/(?<id>[0-9][0-9])/(.*)">
HLSHttpStreamingEnabled true
HLSMediaFileDuration 8000
HttpStreamingContentPath    "${VIDEO_ROOT}/%{env:MATCH_id}"
HLSFmsDirPath ".."
HLSJITConfAllowed true
HLSAMFToId3MappingFilePath conf/AMF_to_Id3_Mapping.xml

The issue is that that with this in place apache won't start - it immediately crashes. I remove %{env:MATCH_id} it loads.

As far as I can tell from the apache docs though that should be valid. So I'm not sure why its dying. Can anyone help?

Apache Docs: http://httpd.apache.org/docs/current/mod/core.html#locationmatch

Thanks!

We're switching from using SQL auth to windows auth per microsoft best practices (https://support.microsoft.com/en-ca/help/2028697/recommendations-and-guidelines-on-configuring-the-authentication-mode)

We've switched the majority of our systems over to no ill effect however when we switched our mobile website over we were fine for several hours up until our site got busy.

At that point connections started queuing, CPU climbs on our database server and load time on the site shoots through the roof.

Normally I would open a case with Micosoft support to troubleshoot but in this instance we're running SQL 2005 which is EoL. (Before lecturing me on SQL 2005 being end of life - we know we're in the process of building SQL 2017 servers and migrating.)

Before we migrate though we wanted to complete the migration from SQL auth to windows auth except for this problem.

Our connection strings to the server look as such:

Provider=SQLOLEDB;Data Source=OURSERVER;Initial Catalog=OURDATABASE;Integrated Security=SSPI;Persist security info=False;Trusted_Connection=Yes;

Data Source=OURSERVER;Initial Catalog=OURDATABASE;Integrated Security=SSPI;

I don't see anything all that interesting in the system, application or security logs either on the database or web server. Likewise I checked our domain controllers. I should also note that we have similar servers in test where we don't see this problem (even during load tests).

Has anyone else experienced anything like this and have any advise for things we might check from a perfmon counters standpoint or otherwise.

I have chef-client installed on a windows node and if I run:

get-windows service *chef* 

I see the service is started. Yet my recipes don't seem to be run unless I open a command prompt or powershell prompt and manually type in:

chef-client 

Obviously that's not what we want - it should run by itself.

Given the service is installed and running how can we troubleshoot why its not working?

Thanks Brad

I need to grab a single file from bitbucket (git) using powershell. Complicating matters I need to do so over http/https as the system I'm running powershell from does not have direct internet access (therefore ssh isn't an option). How can I do so?

I thought that perhaps I could grab it using Invoke-WebRequest and accessing the raw URL. For example: https://bitbucket.org/company/project/raw/HEAD/some/special/file.txt

But it appears to access that you have to authenticate. I've tried setting up authentication but I can't get it working.

I thought next I could just pull it from git using git archive: git archive --remote=git://bitbucket.org/Company/project.git HEAD:some/path/file.txt but that also doesn't seem to work.

So how can I get a single file. I don't want to just do git clone because then I have to pull down (and store) 500MB worth of other files I don't care about in the repo which is an issue for two reasons

1) I have to have space to store all that data (at least temporarily)
2) Unless I want to tie up that storage permanently I have to re-download it each time I need the file. 

The file I need is 10KB so to tie up 500MB for a 10kb file seems absurd.

What's the answer? There's got to be a way to get Invoke-WebRequest to authenticate properly or just download a single file? Can any git gurus - particularly with powershell/windows backgrounds help?

Thanks! Brad

I have a wordpress installation with a large number (millions) of images. I want to optimize those images using ImageMagick mogrify but I only want to process them once (processing them over and over again would result in image quality degradation).

I have to optimize in place (I can't copy the optimized images to a new location because if I did - they wouldn't be accessible to wordpress).

Users can upload new images at any point (even while the optimization process is running) Even more complicated users can upload the same image with the same name. So.... what I'm trying to do is script the following:

• Store the current date/time to a variable
• Get the date the image optimization last ran from a lastrun file. If a last run file doesn't exist default to "1900-01-01 01:01:01"
• Mount a share to a backup server (where unoptimized originals will be stored just in case)
• Search for any files with a *.jpg extension, larger than 10KB, modified any more recently than the last run date
• For any matching files:
  • Copy them to the backup/originals share
  • Optimize them in place:
    • Reduce the quality to 70%
    • Strip any comments and meta data
    • Create progressive jpgs
• Store the current date/time (from the firs step) to the last run time file (this is so we don't re-process the same images over and over again)

The problem with this approach is that all the new images will have a last modified date AFTER the current date/time variable since that's stored at the beginning of the script running.

I could write the date/time the script finished to the lastrun file instead, but then I potentially miss any files upload while the script was running.

So how do I ensure that I don't process any image twice AND I don't miss processing any images.

Here's my script as is:

 # Only uncomment this the very first time to generate the .lastrun file.
 #echo "1900-01-01 01:01:01" >/scripts/config/image-optimizer/.lastrun
 sudo mount.cifs //backup-server/original-images /mnt/originals -v -o user=myuser,dom=mydomain,password=redacted
 last=$(cat /scripts/config/image-optimizer/.lastrun)
 curr=$(date "+%Y-%m-%d %H:%M:%S")
 cd /path/to/wordpress
 find ./ -type f -name "*.jpg" -size +10k -newermt "$last" -exec cp --parents {} /mnt/originals \; -exec mogrify -quality 70 -strip -interlace Plane -monitor {} \;
 echo "$curr" >/scripts/config/image-optimizer/.lastrun

We have three redis 3.2.7 servers. Each redis server runs redis itself and the redis sentinel (for failover). Here's the sentinel config for each node

Slave 1 (sentinel) 192.168.0.15

bind 192.168.0.15
port 26379
logfile "/var/log/sentinel.log"
sentinel myid 57ge72174330d523tcec8adcd1e039c7f51c298c
sentinel monitor mymaster 192.168.0.16 6379 2
sentinel down-after-milliseconds mymaster 5000
sentinel auth-pass mymaster super-secret-password
sentinel config-epoch mymaster 77
dir "/"
sentinel leader-epoch mymaster 0
sentinel known-sentinel mymaster 192.168.0.17 26379 2b72876887f0e2c8db75e6d407babb60eb1a7185
sentinel known-sentinel mymaster 192.168.0.16 26379 06b84b7e6e320797ef49681c2e0598364b673c15
sentinel current-epoch 78

Master (sentinel) - 192.168.0.16

bind 192.168.0.16
port 26379
logfile "/var/log/sentinel.log"
sentinel myid 06b84b7e6e320797ef49681c2e0598364b673c15
sentinel monitor mymaster 192.168.0.16 6379 2
sentinel down-after-milliseconds mymaster 5000
sentinel auth-pass mymaster super-secret-password
dir "/"
sentinel config-epoch mymaster 77
sentinel leader-epoch mymaster 0
sentinel known-sentinel mymaster 192.168.0.17 26379 2b72876887f0e2c8db75e6d407babb60eb1a7185
sentinel known-sentinel mymaster 192.168.0.15 26379 57ge72174330d523tcec8adcd1e039c7f51c298c
sentinel current-epoch 78

Slave S (sentinel) 192.168.0.17

bind 192.168.0.17
port 26379
logfile "/var/log/sentinel.log"
sentinel myid 2b72876887f0e2c8db75e6d407babb60eb1a7185
sentinel monitor mymaster 192.168.0.16 6379 2
sentinel down-after-milliseconds mymaster 5000
sentinel auth-pass mymaster super-secret-password
dir "/"
sentinel config-epoch mymaster 77
sentinel leader-epoch mymaster 0
sentinel known-sentinel mymaster 192.168.0.16 26379 06b84b7e6e320797ef49681c2e0598364b673c15
sentinel known-sentinel mymaster 192.168.0.15 26379 57ge72174330d523tcec8adcd1e039c7f51c298c
sentinel current-epoch 78

If I stop the redis-server and redis-sentinel on the Master node I see the following on the slaves:

2664:X 01 Mar 16:17:15.563 # +sdown master mymaster 192.168.0.16 6379

So the sentinels are seeing the Master as down. However no failover occurs. If I check the slaves using:

redis-cli -h 127.0.0.1 -p 26379 -a super-secret-password sentinel get-master-addr-by-name mymaster

I get:

1) "192.168.0.16"
2) "6379"

So they're seeing the old Master (which has been shutdown as still their Master. If they're seeing the Master as down why aren't they electing a new Master?

By the way I believe this configuration was working when we were running Redis 2.8.9. So I can only guess that redis sentinel might have changed in 3.2.7. I'm just not sure where to begin looking.

Any help or pointers would be appreciated.

Thanks Brad

We're trying to use IIS shared configuration on Server Core 2012 R2.

It works perfectly with one exception which is that if the SMB share that stores the IIS config files goes offline IIS crashes and won't automatically recover even if the SMB share comes back online.

The only solution I can find to this issue is this article (https://www.iis.net/learn/web-hosting/configuring-servers-in-the-windows-web-platform/offline-files-for-shared-configuration) which describes setting up offline files.

The problem with that is that offline files is part of the desktop-experience feature which requires "Server Graphical Shell" and "Graphical Tools and Infrastructure". So to utilize that we'd need to basically convert our server core boxes to the full version of windows. Not exactly what we want - we're using server core to minimize the footprint of our servers.

I'm trying to figure out if there is some alternative we can use. I've found several different solutions which will sync content from the SMB share to the local PC (GoodSync, SyncBack Pro, Sync Toy, etc) - that's no problem.

The issue is that offline files "intercepts the API call to access the files on the file server, and get the data in the local device cache". So it transparently takes care of redirecting the request from the remote share to local storage. None of the alternatives I can find to offline files handle this piece.

Does anyone know of any syncing tools that would transparently handle redirecting requests to the local filesystem when the smb share is unavailable?

We have a number of windows 2012 server core systems with powershell setup as the default shell using the following commands:

$RegPath = "Microsoft.PowerShell.Core\Registry::HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\winlogon"
Set-ItemProperty -Confirm  -Path $RegPath -Name Shell -Value 'cmd.exe /C start /max PowerShell.exe -noExit'

I've figured out that we can customize the powershell font color with a special powershell script at c:\Windows\System32\WindowsPowerShell\v1.0\profile.ps1. This script gets used by all users.

However now I want to customize the font face and font size (again for all users) that's persistent. I.E. If I log out of the server and log back in I want the settings to be retained. Likewise if I login as administrator, or my own account powershell should look identical - use the same font color, font face and font size.

With Powershell ISE it seems possible to set the font face and font size using:

$psISE.Options.FontName = 'Lucida Sans Console' 
$psISE.Options.FontSize = 14

Whats the equivalent for powershell itself though?

I have snmpd executing a script under /custom/location/myscript.sh. If I disable selinux all works well. If I enable selinux it stops working and if I do an audit2why I see the following:

type=AVC msg=audit(1371990300.064:23445): avc:  denied  { read open } for  pid=21423 comm="snmpd" name="myscript.sh" dev=dm-0 ino=1046532 scontext=unconfined_u:system_r:snmpd_t:s0 tcontext=unconfined_u:object_r:default_t:s0 tclass=file

    Was caused by:
            Missing type enforcement (TE) allow rule.

            You can use audit2allow to generate a loadable module to allow this access.

type=AVC msg=audit(1371990313.066:23446): avc:  denied  { read open } for  pid=21425 comm="snmpd" name="myscript.sh" dev=dm-0 ino=1046532 scontext=unconfined_u:system_r:snmpd_t:s0 tcontext=unconfined_u:object_r:default_t:s0 tclass=file

    Was caused by:
            Missing type enforcement (TE) allow rule.

            You can use audit2allow to generate a loadable module to allow this access.

If I do an ls -Z /custom/location I see the following:

-rwxr-xr-x. root root unconfined_u:object_r:default_t:s0 myscript.sh

So I need to do an chcon-R on the directory. I tried:

chcon -R -u unconfined_u -r system_r -t snmpd_t /custom/location

Only to get the following error:

chcon: failed to change context of `myscript.sh' to `unconfined_u:system_r:snmpd_t:s0': Permission denied

I also tried using audit2allow to create a custom policy but that didn't seem to work.

I really don't want to disable selinux. Can anyone assist me in understanding how I can allow snmpd the ability to read from /custom/location?

Thanks Brad

Since the output was so long I am truncating a bit - but here's the results of ausearch:

----
time->Sun Jun 23 08:27:13 2013
type=SYSCALL msg=audit(1371990433.731:23453): arch=c000003e syscall=189 success=no            
exit=-13 a0=11760e0 a1=3e50616319 a2=1178ef0 a3=21 items=0 ppid=20865 pid=21434 auid=0 
uid=0 gid=0 euid=0 suid=0 fsuid=0 egid=0 sgid=0 fsgid=0 tty=pts0 ses=3863 comm="chcon"
exe="/usr/bin/chcon" subj=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 key=   
(null)

type=AVC msg=audit(1371990433.731:23453): avc:  denied  { relabelto } for  pid=21434  
comm="chcon" name="scripts" dev=dm-0 ino=1046530 
scontext=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 
tcontext=unconfined_u:system_r:snmpd_t:s0 tclass=dir
----
time->Sun Jun 23 08:27:24 2013
type=SYSCALL msg=audit(1371990444.075:23454): arch=c000003e syscall=59 success=no     
exit=-13 a0=7fdb8933c6d0 a1=7fdb8933c3a0 a2=7fff16ef1790 a3=1f items=0 ppid=18001 
pid=21435 auid=0 uid=0 gid=0 euid=0 suid=0 fsuid=0 egid=0 sgid=0 fsgid=0 tty=(none) 
ses=3765 comm="snmpd" exe="/usr/sbin/snmpd" subj=unconfined_u:system_r:snmpd_t:s0 key=
(null)

type=AVC msg=audit(1371990444.075:23454): avc:  denied  { read open } for  pid=21435 
comm="snmpd" name="snmp.sh" dev=dm-0 ino=1046532 
scontext=unconfined_u:system_r:snmpd_t:s0 tcontext=unconfined_u:object_r:default_t:s0 
tclass=file
----

In our varnish config we have a vcl_fetch function defined as follows:

if (beresp.ttl > 0s) {
   /* Remove Expires from backend, it's not long enough */
   unset beresp.http.expires;

   /* Set the clients TTL on this object */
   set beresp.http.cache-control = "max-age=900";

   /* Set how long Varnish will keep it */
   set beresp.ttl = 1h;

   /* marker for vcl_deliver to reset Age: */
   set beresp.http.magicmarker = "1";
 }

Then in vcl_deliver:

if (resp.http.magicmarker) {
   /* Remove the magic marker */
   unset resp.http.magicmarker;

   /* By definition we have a fresh object */
   set resp.http.age = "0";
}

The issue is that if I setup a script (either a shell script or a php script) to hit the same URL, I first see a MISS, followed by a HIT (all good so far).

What I am expecting to sit is repeated HITS until the 1 hour mark has passed, one MISS at the 1 hour mark, followed by an hour of HITs again. In other words because I'm hitting the same URL repeatedly it should cycle between HIT and MISS every 1 HOUR.

Instead what I see is something like this:

 6/20/2013 6:30:05 - STRING HIT NOT FOUND
 6/20/2013 10:32:11 - STRING HIT NOT FOUND
 6/21/2013 12:33:06 - STRING HIT NOT FOUND
 6/21/2013 2:34:06 - STRING HIT NOT FOUND

It's now 5:57AM (almost 4 hours since the HIT not found).

The cache misses above do all appear around the half hour mark but they're 4 hours, 2 hours and 2 hours apart and the TTL is set to 1 hour.

So I don't understand why the object is being kept longer than 1 hour and why its evicted at seemingly random intervals. How can I better determine whats going on?

By the way I am using varnish 3 if that helps.

When I attempt to Secure FTP (using WinSCP) to my CentOS 6.4 box I'm prompted with a message that states:

Received too large (168438314 B) SFTP packet. Max supported packet size is 1024000 B.

This error is typically caused by message printed from startup script (like profile). The message may start with ""\n\n**"" 

Cannot initialize SFTP protocol. Is the host running an SFTP server

This appears to be caused by my /etc/profile.d/custom.sh script which gets run during login. Its setup to output a welcome message, as well as a system summary (CPU/Memory/Disk Usage).

I really need the custom.sh script to run when logging into the local console OR when logging in via SSH. However I don't want it to run when logging into Secure FTP (SFTP).

Is there some way to configure it not to run when logging in via Secure FTP?

Or alternatively if I can't stop it from running is there a way I can detect that its being run during a secure FTP login and not output anything. So in pseudo code something like:

If logged in via secure FTP then output nothing Else if logged in via SSH or Local Console output message of the day (system stats).

Thanks! Brad

UPDATE - There is documentation supporting this issue here: http://winscp.net/eng/docs/requirements#remote_environment

The suggestion is to detect if the session is interactive or not. I tried doing that by adding the following to the top of my custom.sh but its still not working:

if [ -z $PS1 ]; then
### if [ -v PS1 ]   # On Bash 4.2+ ...
    # non-interactive
    return
fi

Then the rest of my script runs outputting the MOTD.

Should SecureFTP be detected as non-interactive or interactive? If its non-interactive why isn't the code above existing the script when I attempt to login via Secure FTP?

I'm mounting an SMB/CIFS share on my CentOS 6.4 server using /etc/fstab as follows:

Mount windows share to /mnt/mywebsite

//10.1.1.4/mywebsite /mnt/mywebsite cifs username=httpd,password=topsecret,uid=apache,gid=apache 0 0

I then issue mount -a and I see that my mount is created and the owner is apache:apache.

I can create files and directories just fine but if I try and do a chmod on files. For example:

mkdir test chmod 755 test

I get the following error: chmod: changing permissions of `test': Permission denied

I'm trying to figure out why I'm getting this. Is there an SELINUX policy blocking me? Do I need to add some additional parameters in my /etc/fstab?

I tried doing grep http /var/log/audit/audit.log | audit2why and I'm not seeing SELINUX blocking me.

Thanks Brad

I'm looking for the easiest and most sane way of installing the latest version of apache (currently 2.4.4) and PHP (currently 5.4.13) on CENTOS 6.4 (x64) and I've run into no end of issues.

If I use yum with the default repositories the latest version available is 2.2.15 (which seems to have been released Mar 2010) 3 years ago. Really? I want a three year old web server on my brand new server?

So I next looked at building from source - which I was able to do but when I finished installing I wanted to test uninstalling so I did a "make uninstall" and was greeted with a message make: *** No rule to make target `uninstall'.

Great so I can't uninstall it in an automated manner to upgrade later on? Additionally when I went to build PHP from source I was faced with dependency after dependency. I proceeded with installing one by one until I eventually gave up. This doesn't seem like a reasonable approach - no option to uninstall - never-ending dependencies.

I next looked at alternative repositories like REMI - it seems REMI offers PHP5.4.13 it requires apache 2.2.15 You'll get a message stating Requires: httpd-mmn = 20051115 (which is apparently part of Apache 2.2.15

I next looked at building my own RPM from source. (At least that way I could easily uninstall them.) That worked fairly well - initially. I was able to easily build an RPM of apache, and with a little effort I was able to tweak PHP to look for apache 2.4.13 but after I do that I'm still prompted with an error that php54 requires Requires: libcurl.so.3()(64bit). I seem to have curl and libcurl, libcurl-devel installed.

At this point I'm frustrated - there has to be an easier way to get an updated Apache and PHP stack without this craziness. How are other admins doing this? Is there some secret way of doing this that I'm not aware of? Or is everyone just running really old versions of Apache and PHP and ignoring that they're riddled with security holes?

I'm a bit new to Memcache and Redis and I'm trying to understand the difference between the two and what each actually is.

As has been explained to me Memcache is a database caching solution. Data from frequently run queries are cached to be re-used at a later point speeding up response time. Is that an accurate description?

What about Redis? Does that work in a similar manner? In other words does it cache data it receives from a backend database? I keep reading that Redis is is a key value store solution which to me sounds like more of a replacement for a database server vs a db caching solution.

Any help understanding what each is and what the differences are would be greatly appreciated.

Thanks Brad

I'm trying to wrap my head around client side (web browser) caching and how it works in relation to IIS 7.5 cache control headers.

In particular:

1. If we want to force clients to reload cached resources, how must IIS be configured?

2. Do we need to set expire web content immediately if the resources on the server have a more recent Modified Date (or ETag value)?

3. Right now we're not setting any cache headers. So if I set a cache header of no-cache (which I think is the equivalent of expire web content immediately) will that force the web browser to obtain a new version of a particular file. Or will the browser only request a new version after it deems its current copy to be stale and then from that point forward not cache it?

Would a best practice be to set a cache control flag of 1 week, then 8 days before I know I am going to make a change set the cache control down to for instance 30 minutes?

But if I do that and then need to immediately expire an item from users caches because there was an issue with it how do I do that?

I've configured a WAMP (Windows Apache MySQL and PHP) stack when when configured to use local storage takes 3-4 seconds to load. When I use an SMB/UNC share it takes 12-15 seconds to load.

Here are the two lines in my httpd.conf:

 #DocumentRoot "//10.99.108.11/test_htdocs"
 #<Directory "//10.99.108.11/test_htdocs">

 #DocumentRoot "C:/www"
 #<Directory "C:/www">

Is there performance tuning I can do on windows server 2008 R2 to improve performance or is there another way to improve performance using smb