golemwashere's questions

I am using this curl:

curl -vvv --url 'smtp://smtp.office365.com:587' \
--ssl-reqd \
--mail-from "$user" \
--mail-rcpt "$destuser" \
--upload-file ./mail.txt \
--oauth2-bearer $accesstoken

to try to send an email from an account where I registered an oauth application with these Graph Permission - Delegated Permissions

SMTP.Send
Mail.Read
Mail.Send
openid
offline_access

but I always get error: 530 5.7.57 Client not authenticated to send mail.

both the user mailbox and the transport have SmtpClientAuthenticationDisabled to $false

I am quite confident that the access token is valid, I also tested the refresh token which works, any idea on how to make this work?

Is curl compatible with https://docs.microsoft.com/en-us/exchange/client-developer/legacy-protocols/how-to-authenticate-an-imap-pop-smtp-application-by-using-oauth

?

Can anyone point to any public doc stating that pools from ntp.org and time.windows.com are safe from next thrusday gps week rollover ?

I am looking to something similar to this Red Hat KB

https://access.redhat.com/solutions/4018911

thanks g

I have an apache 2.2.29-1.4 with mod_security 2.8.0-5.25 which is a reverse proxy with mod_proxy_http for a local java application.

I have false positive on some urls and would like to whitelist some OWASP rules just on the given URIs.

Currently globally whitelisting a rule with

SecRuleRemoveById 960010

works perfectly

But both

<LocationMatch ^/(myapp/mymethod.do.*)$>
SecRuleRemoveById 960010
  ProxyPassMatch http://localhost:8080/$1
</LocationMatch>

and

<Location "/myapp/mymethod.do">
SecRuleRemoveById 960010
</Location>
ProxyPass /myapp/mymethod.do http://localhost:8080/myapp/mymethod.do
ProxyPassReverse /myapp/mymethod.do http://localhost:8080/myapp/mymethod.do

do not whitelist the rule.

I also tried adding new mod_security rules to modify the existing ones but whitelisting an URI always fails...

Any insight?

I am trying to setup IIS 7.0 dispatcher module for CQ5 following the setup: http://dev.day.com/docs/en/cq/current/deploying/dispatcher/disp_install.html.

I am trying on windows 2008 sp2 64bit and a local publishing istance on port 4203.

The dispatcher works as long as cache is disabled, once I enable it, files are written in the document root but the homepage is returned empty and I get in my dispatcher.log:

[Fri Mar 07 16:38:24 2014] [D] [3180(2108)] checking [/content/geometrixx-outdoors/en.html]
[Fri Mar 07 16:38:24 2014] [D] [3180(2108)] never flushed [E:\wwwroot2\.stat] -> use cache [E:\wwwroot2\content\geometrixx-outdoors\en.html]
[Fri Mar 07 16:38:24 2014] [D] [3180(2108)] cache-action for [/content/geometrixx-outdoors/en.html]: SPOOL
[Fri Mar 07 16:38:24 2014] [D] [3180(2108)] request declined
[Fri Mar 07 16:38:24 2014] [I] [3180(2108)] "GET /content/geometrixx-outdoors/en.html" 0 - 0ms
[Fri Mar 07 16:38:24 2014] [W] [3180(2108)] HSE_REQ_EXEC_URL(/content/geometrixx-outdoors/en.html) returned: 000003E9

I get a code 200 from my IIS 7.0

I gave modify permission to IUSER and tried also a

icacls e:\wwwroot2 /grant "IIS APPPOOL\DefaultAppPool":(OI)(CI)(RX)

(should not be necessary since I am on 7.0 and not 7.5) The dispatcher is installed as handler mapping

if I bypass cache with a query string as in

http://myserver/content/geometrixx-outdoors/en.html?a=1

it works

any hint?

I have a postfix installation with amavis and postfwd. I'd like to block a single sender only for a single receiver. I tried with header_checks but multiline checking I was thinking something like

id=BLOCKIT
[email protected]; [email protected]; action=REJECT no sender

on postfwd but I can't find the proper syntax... Has anyone some suggestions?

we have some outgoing mail server with latest postfix and amavis.

I would need to intercept some outgoing email for some given senders and forward them to a "logging" account + the regular receivers.

From what I read looks like this job could be only done by a custom filter, since we already have amavis in place I was wondering if this could be done directly by amavis.

Any custom postfix filter example with filtering by sender would be welcome too.

Thanks

I would like to use postfwd version 2 to limit the amount of daily mail sent by my sasl authenticated users.

I installed latest tarball: postfwd-1.35 with latest postfix from Centos 6.4

In my I have only this rule

id=RULEZEROSASL
  sasl_username=~/^(\S+)$/
  action=rcpt(sasl_username/500/86400/REJECT only 500 recipients per day for $$sasl_username) 

which should reject only mail with authenticated users (not mail from trusted mailservers).

My postfwd2 listens on tcp 10045 and in my postfix main.cf I have

# Restriction Classes
smtpd_restriction_classes       = postfwdcheck
postfwdcheck                    = check_policy_service inet:127.0.0.1:10045
127.0.0.1:10045_time_limit      = 3600

...

smtpd_recipient_restrictions =
        permit_mynetworks
        permit_sasl_authenticated
        permit_tls_clientcerts
        reject_unauth_destination
        check_recipient_access  hash:/etc/postfix/access
        reject_invalid_helo_hostname
# postfwd con rate limiting
        check_policy_service inet:127.0.0.1:10045
        warn_if_reject reject_non_fqdn_helo_hostname
        warn_if_reject reject_unknown_helo_hostname
        warn_if_reject reject_unknown_client
        reject_non_fqdn_sender
        reject_non_fqdn_recipient
        reject_unknown_sender_domain
        reject_unknown_recipient_domain
        warn_if_reject reject_unverified_sender
        reject_unverified_recipient
        reject_rbl_client zen.spamhaus.org
        permit  

in /etc/postfix/policy

.   postfwdcheck

I see no rule matching entries in log and the command

postfwd2 -vv --dumpcache -f /etc/postfwd.cf

shows the request number

[STATS] postfwd2::policy 1.35: **5** requests since 0 days, 01:05:31 hours

increasing only for manual tests done with:

 nc 127.0.0.1 10045 <request.sample

Any idea why postfwd is not engaged by postfix?

I am trying to replace an existing qmail installation with postfix.

Qmail is currently authenticating outgoing smtp from users clients with a perl script. This script checks user's provided username+password performing a AUTH LOGIN check on another external smtp server.

If the remote server responds with 235 return code, the provided username/password are ok and user is authenticated for qmail as well.

From what I understand it is possibile to use Cyrus or Dovecot SASL libraries on postfix to authenticate users, is it possibile to configure saslauthd or any of those libraries to replicate the perl script behaviour?

Unluckyly I do not have any other kind of access to the external smtp server cointaining users and passwords, I only can test AUTH LOGIN on an smtp connection.

Thanks

I am trying to use httperf to stress test a server using the following syntax

    httperf --hog --client=0/1 --server myserver.com \
--wsesslog=2000,2,mysessions.cfg --max-piped-calls 5 --rate 150 

this should create 2000 sessions at a rate of 150 req/s

mysessions.cfg contains

/path1/url think=2.0
/common/css/base.css
/images/image.png

..

My problem is I would like to specify a burst-lenght (concurrent calls inside session) greater than 2 but I always get

Maximum connect burst length: 2

the ­­burst­length=N is not usable with wsessionlog and I didn't see any way to specify it inside mysessions.cfg

Any insight?

I have access to Redhat linux 5.7 guests on vmware ESX (probably 4) hosts which I suspect is often overloaded ( CPU and IO ).

I have no visibility on the ESX host which is managed by another company.

Is it possible to get some information on host state by vmware agent?

Which are the best OS parameter to monitor to get info on host state?

We are experiencing random blank email bodies on email sent between internal users on Exchange 2010 sp1 rollup pack 3 and clients using outlook 2010 sp 1 on windows 7 with latest patches. The same email sent to different users is correctly received by everyone but one single user (it happens randomly on different users). In outlook the user can see headers (Sender, Receiver, subject, date, etc) but the body is completely blank.

We already tried with no luck:

• disabling antivirus.

• setting outlook email compose setting to rtf.

• deleting and recreating profiles and also disable cached mode on profiles.

Any suggestion would be welcome.

I managed to get HAproxy working in a failover configuration as suggested on the architecture.txt document using keepalived.

I used CentOS 5.5, HAproxy rpm (haproxy-1.3.25-1.el5) from epel repo, and keepalived rpm (keepalived-1.1.15-0.el5.centos) from CentOS testing repo.

This is my keepalived configuration

vrrp_script chk_haproxy {           # Requires keepalived-1.1.13
    script "killall -0 haproxy"     # cheaper than pidof
    interval 2                      # check every 2 seconds
    weight 2                        # add 2 points of prio if OK
}

vrrp_instance VI_1 {
    interface eth0
    state MASTER
    virtual_router_id 51
    priority 101                    # 101 on master, 100 on backup
    virtual_ipaddress {
    10.0.0.30            
    10.0.0.31            
    }
    track_script {
        chk_haproxy
    }
}

What I would like now is having my backup node being able to take over the actual connections open when the master is failing.

I heard this can be done in multiple ways (conntrackd), using LVS features, etc... I'd like to have suggestions on the simplest and more robust way to achieve it, using available rpms would be best.

Thanks G

I am trying to install a Centos 5.5 xen domU using this method: http://wiki.centos.org/HowTos/Xen/InstallingCentOSDomU

It suggests a kickstart config file served by the local apache. I managed to install it on a physical machine , but when I am trying the installation on a virtualbox instance (dom0 on latest virtualbox) anaconda exits with the following errors:

Running anaconda, the CentOS system installer - please wait...                 
Traceback (most recent call last):
  File "/usr/bin/anaconda", line 733, in ?
    vncksdata = setVNCFromKickstart(opts)
  File "/usr/bin/anaconda", line 277, in setVNCFromKickstart
    ksparser.readKickstart(opts.ksfile)
  File "/usr/lib/python2.4/site-packages/pykickstart/parser.py", line 1209, in readKickstart
    self.handleCommand(lineno, args)
  File "/usr/lib/python2.4/site-packages/pykickstart/parser.py", line 1069, in handleCommand
    raise KickstartParseError, formatErrorMsg(lineno, msg=_("Unknown command: %s" % cmd))
pykickstart.parser.KickstartParseError: The following problem occurred on line 1 of the kickstart file:

Unknown command: <!DOCTYPE

install exited abnormally [1/1] 

an interactive Installation works and I managed to get a funcional xen vm, the kickstart file is downloable via links, I have Selinux disabled. I am pretty sure of the kickstart file syntax, has anyone any hint?

I've been asked to size a web architecture for an excessive number of concurrent users ( hundreds of thousands ).

I'm having a hard time convincing these people that unless you are in the top 5 of your country websites it's quite hard to hit those numbers.

Can anyone provide some real world case studies providing stats for total / concurrent users explaining what is the usual ratio between total vs concurrent?

I have a dell poweredge 2970 server with a serial console over ip enabled on the first ethernet.

The server with Red Hat enteprise 5.4 is configured to have console over ttyS1 in grub (console=ttyS1,57600 ) and in /etc/inttab I have:

S1:2345:respawn:/sbin/agetty -L -h ttyS1 57600 vt100

I can connect to console from another server using :

ipmitool -I lanplus -H myipmiIPaddress -U root -P password sol activate

but in the server I have /var/log/messages flooded with init: Id "S1" respawning too fast: disabled for 5 minutes.

If I remove the entry in inttab I can't connect anymore so I suppose it's not completely wrong, how could I avoid it respawing?

We have a Sun Netra that someone booted with a monitor connected to the VGA output and the console went to screen. I'd like to switch that to serial without a reboot, I tried

eeprom output-device=ttya
eeprom input-device=ttya

which is accepted but not enough, I am afraid it's still on screen. Is there any mean to switch it at runtime without rebooting?

I am trying to execute a script through a SNMP GET using Net-Snmp included in the latest Solaris 10 (10/09 s10s_u8wos_08a SPARC).

I have a non-global zone which runs NET-SNMP version 5.0.9 patched to the last SMA patch (120272-26).

The zone responds to a snmpwalk but if I add something like

exec 1.3.6.1.4.1.2021.8.1.100 test /bin/echo "hello golem" 

in /etc/snmp/conf/snmpd.conf and restart:

svcadm restart sma

if I try to walk .1.3.6.1.4.1.2021.8 (UCD-SNMP-MIB::extTable) I get no objects.

In other similar cases I solved installing the version from sunfreeware.com but it looks like the current version (netsnmp-5.4.2.1-sol10-sparc-local) is not working inside a zone.

Does anyone know how to enable the extTable tree? What is really strange is that the man page for snmpd.conf shows the exec command and the demon says it's capable to understand exec:

 prompt# /usr/sfw/sbin/snmpd -H 2>&1| grep exec
        exec                     [miboid] name program arguments

PS I get no hints from /var/log/snmpd.log

I have an application (CQ5) that requires enabling unauthenticatedIdentity on jbossdir/conf/login-config.xml I used:

   <authentication>
      <login-module code = "org.jboss.security.auth.spi.UsersRolesLoginModule"
         flag = "required" >
      <module-option name="unauthenticatedIdentity">nobody</module-option>
    </login-module>
   </authentication>

then I tried to copy jbossdir/conf/props/jmx-console-users.properties,jmx-console-roles.properties into users.properties and roles.properies (same dir).

I still get this error:

ERROR [org.jboss.security.auth.spi.UsersRolesLoginModule] Failed to load users/passwords/role files java.io.IOException: No properties file: users.properties or defaults: defaultUsers.properties found

where should I put those files?