Re: [cas-user] shib idp 3 CAS support

2015-07-18 Thread Dmitriy Kopylenko
As the matter of fact we have done so :-) The implementation needs some 
polishing before it could be publicly announced (on the shib lists), but it's 
coming :-)

Have a great weekend. 

D. 

Sent from my iPhone

 On Jul 18, 2015, at 15:58, Paul B. Henson hen...@cpp.edu wrote:
 
 On Fri, Jul 17, 2015 at 03:25:35PM -0400, Dmitriy Kopylenko wrote:
 Just want to conclude this thread with a pretty good read about Hazelcast
 
 So... Any chance you guys at Unicon have any interest in putting together a
 Hazelcast based clustering backend for idp 3 :)?
 
 
 -- 
 Paul B. Henson  |  (909) 979-6361  |  http://www.cpp.edu/~henson/
 Operating Systems and Network Analyst  |  hen...@cpp.edu
 California State Polytechnic University  |  Pomona CA 91768
 
 -- 
 You are currently subscribed to cas-user@lists.jasig.org as: 
 dkopyle...@unicon.net
 To unsubscribe, change settings or access archives, see 
 http://www.ja-sig.org/wiki/display/JSG/cas-user

-- 
You are currently subscribed to cas-user@lists.jasig.org as: 
arch...@mail-archive.com
To unsubscribe, change settings or access archives, see 
http://www.ja-sig.org/wiki/display/JSG/cas-user



Re: [cas-user] shib idp 3 CAS support

2015-07-18 Thread Paul B. Henson
On Fri, Jul 17, 2015 at 03:25:35PM -0400, Dmitriy Kopylenko wrote:
 Just want to conclude this thread with a pretty good read about Hazelcast

So... Any chance you guys at Unicon have any interest in putting together a
Hazelcast based clustering backend for idp 3 :)?


-- 
Paul B. Henson  |  (909) 979-6361  |  http://www.cpp.edu/~henson/
Operating Systems and Network Analyst  |  hen...@cpp.edu
California State Polytechnic University  |  Pomona CA 91768

-- 
You are currently subscribed to cas-user@lists.jasig.org as: 
arch...@mail-archive.com
To unsubscribe, change settings or access archives, see 
http://www.ja-sig.org/wiki/display/JSG/cas-user


Re: [cas-user] shib idp 3 CAS support

2015-07-18 Thread Paul B. Henson
On Sat, Jul 18, 2015 at 04:29:55PM -0400, Dmitriy Kopylenko wrote:
 As the matter of fact we have done so :-) The implementation needs
 some polishing before it could be publicly announced (on the shib
 lists), but it's coming :-)

Sweet. If it's not out before I get around to testing idp 3 I might bug
you to beta test :).

Thanks...


-- 
Paul B. Henson  |  (909) 979-6361  |  http://www.cpp.edu/~henson/
Operating Systems and Network Analyst  |  hen...@cpp.edu
California State Polytechnic University  |  Pomona CA 91768

-- 
You are currently subscribed to cas-user@lists.jasig.org as: 
arch...@mail-archive.com
To unsubscribe, change settings or access archives, see 
http://www.ja-sig.org/wiki/display/JSG/cas-user


Re: [cas-user] shib idp 3 CAS support

2015-07-17 Thread Dmitriy Kopylenko
Just want to conclude this thread with a pretty good read about Hazelcast. 
Arguably, by now Hazelcast software (at v3.5) is a pretty robust, 
well-implement and mature distributed grid/cache.

http://www.brickendon.com/articles/achieving-low-latency-using-distributed-cache/
 
http://www.brickendon.com/articles/achieving-low-latency-using-distributed-cache/

Best,
D.

 On Jul 14, 2015, at 3:32 PM, Paul B. Henson hen...@cpp.edu wrote:
 
 From: Marvin Addison
 Sent: Tuesday, July 14, 2015 6:33 AM
 
 Correct. What makes this acceptable in many if not most cases is that the 
 lost
 state is SSO state where the effect on the user is to log in again. As 
 failure
 modes go, that's graceful behavior.
 
 Arguably true, but still not optimal :). Contrary to what seems to be the 
 average deployment, I also encrypt the cluster replication traffic over the 
 wire, so I perhaps have stricter requirements for perfection than generally 
 considered ;).
 
 Peer-to-peer replication incurs a cost and in my experience the failure 
 modes of
 replication are orders of magnitude worse than anything I've seen with
 memcached. Perhaps over time Hazelcast will prove itself both reliable and 
 fault
 tolerant, but it's patently new technology at this point and needs some road
 time to convince me.
 
 I load tested it pretty heavily including random node restarts and it never 
 blipped. We've been running it in production for about a year and a half and 
 I haven't seen a single problem (knock on wood). We've probably done at least 
 4-5 rolling updates since then where we pulled a node out of the cluster and 
 then stuck it back in, I'm unaware of any user facing issues or unnecessary 
 re-authentications. In any case, I'm pretty happy with it :), and wouldn't 
 really want to trade it out for memcached.
 
 Thanks…
 
 --
 Paul B. Henson  |  (909) 979-6361  |  http://www.cpp.edu/~henson/
 Operating Systems and Network Analyst  |  hen...@cpp.edu
 California State Polytechnic University  |  Pomona CA 91768
 
 
 
 -- 
 You are currently subscribed to cas-user@lists.jasig.org as: 
 dkopyle...@unicon.net
 To unsubscribe, change settings or access archives, see 
 http://www.ja-sig.org/wiki/display/JSG/cas-user


-- 
You are currently subscribed to cas-user@lists.jasig.org as: 
arch...@mail-archive.com
To unsubscribe, change settings or access archives, see 
http://www.ja-sig.org/wiki/display/JSG/cas-user

Re: [cas-user] shib idp 3 CAS support

2015-07-14 Thread Marvin Addison

 The primary reason I did not like the memcached implementation was that it
 is not truly fault-tolerant. As I understand it, any given key is hashed
 and stored on one of the available memcached nodes. A client that needs it
 performs the same hash, and tries to retrieve it from the same node. I have
 a three node CAS cluster. With the memcached ticket registry, if one of the
 nodes were to crash, I would simply lose a third of my state.


Correct. What makes this acceptable in many if not most cases is that the
lost state is SSO state where the effect on the user is to log in again. As
failure modes go, that's graceful behavior.


 On the other hand, with the hazelcast ticket registry backend, a backup
 copy of the state on any given node is available on another node, and if
 you lose only a single node, you lose no state.


Peer-to-peer replication incurs a cost and in my experience the failure
modes of replication are orders of magnitude worse than anything I've seen
with memcached. Perhaps over time Hazelcast will prove itself both reliable
and fault tolerant, but it's patently new technology at this point and
needs some road time to convince me.

Best,
M

-- 
You are currently subscribed to cas-user@lists.jasig.org as: 
arch...@mail-archive.com
To unsubscribe, change settings or access archives, see 
http://www.ja-sig.org/wiki/display/JSG/cas-user


RE: [cas-user] shib idp 3 CAS support

2015-07-14 Thread Paul B. Henson
 From: Marvin Addison
 Sent: Tuesday, July 14, 2015 6:33 AM
 
 Correct. What makes this acceptable in many if not most cases is that the lost
 state is SSO state where the effect on the user is to log in again. As failure
 modes go, that's graceful behavior.

Arguably true, but still not optimal :). Contrary to what seems to be the 
average deployment, I also encrypt the cluster replication traffic over the 
wire, so I perhaps have stricter requirements for perfection than generally 
considered ;).

 Peer-to-peer replication incurs a cost and in my experience the failure modes 
 of
 replication are orders of magnitude worse than anything I've seen with
 memcached. Perhaps over time Hazelcast will prove itself both reliable and 
 fault
 tolerant, but it's patently new technology at this point and needs some road
 time to convince me.

I load tested it pretty heavily including random node restarts and it never 
blipped. We've been running it in production for about a year and a half and I 
haven't seen a single problem (knock on wood). We've probably done at least 4-5 
rolling updates since then where we pulled a node out of the cluster and then 
stuck it back in, I'm unaware of any user facing issues or unnecessary 
re-authentications. In any case, I'm pretty happy with it :), and wouldn't 
really want to trade it out for memcached.

Thanks…

--
Paul B. Henson  |  (909) 979-6361  |  http://www.cpp.edu/~henson/
Operating Systems and Network Analyst  |  hen...@cpp.edu
California State Polytechnic University  |  Pomona CA 91768



-- 
You are currently subscribed to cas-user@lists.jasig.org as: 
arch...@mail-archive.com
To unsubscribe, change settings or access archives, see 
http://www.ja-sig.org/wiki/display/JSG/cas-user

Re: [cas-user] shib idp 3 CAS support

2015-07-13 Thread Marvin Addison

 Would the idp multifactor mechanisms be usable for CAS clients?


Yes, though it may be better to discuss further on the related shib-users
thread.


 I don't
 particularly like the memcached backend, which looks like the only
 current option (other than a database) to cluster idp 3.


I'd be curious to know why you don't like it. I encourage you to discuss
further on shib-users.

Yes, from what I've seen the attribute filter basically functions as the
 service registry for CAS clients, but gives you the idp feature of only
 allowing certain values of some attributes rather than all.


The ability to mutate attributes is a notable capability of the IdP
attribute engine.


 I need to decide
 whether to keep a separate CAS deployment ... or migrate CAS clients to
 the idp CAS support.


I was in the same situation, so I'll share my decision FWIW. I'm a single
FTE running two production systems, Shib IdP and Jasig CAS server, that
have the same purpose but with complementary capabilities. It made all the
sense in the world to consolidate onto a single platform that could provide
everything we needed. Fortunately I had the benefit of developing the CAS
protocol support in the IdP expressly to meet our requirements. Thus we had
assurance our needs would be met by a single platform and it became a
straightforward economic decision: consolidate and save. Of course our
needs and yours are likely different; you'll have to evaluate for yourself
whether a single platform meets your requirements.

M http://www.ja-sig.org/wiki/display/JSG/cas-user

-- 
You are currently subscribed to cas-user@lists.jasig.org as: 
arch...@mail-archive.com
To unsubscribe, change settings or access archives, see 
http://www.ja-sig.org/wiki/display/JSG/cas-user


Re: [cas-user] shib idp 3 CAS support

2015-07-12 Thread Paul B. Henson
On Sun, Jul 12, 2015 at 11:54:34PM +, Marvin Addison wrote:

 I'm very familiar with both from a development and deployment perspective.
 Much of the growth of Jasig CAS over the past several years has been in the
 area of multi-protocol support, so you'll lose anything that's not CAS or
 SAML; OAuth protocol support comes to mind. That's probably the biggest
 functional area of distinction.

We don't currently use OAuth, and I don't think we have any short to
mid-term plans to. We are going to be looking at multifactor
authentication soon though. I'm not that familiar with them, but I know
both CAS and shib idp have mechanisms for integrating multifactor. Would
the idp multifactor mechanisms be usable for CAS clients?

 I would say Jasig CAS has more mature management capabilities; the service
 manager UI comes to mind immediately.

I actually use the Unicon JSON backend in read-only mode, the config file
itself is stored under version control. So no worries there :).

 and effort before the IdP catches up in that area. There are also a lot
 more integration options for Jasig CAS, mostly in the area of storage
 backends like Ehcache and Hazelcast. Again, it's easy to develop these

Yeah, I gotta say I love the Hazelcast ticket registry. After fighting
with the ehcache one for a month or so it was much easier to get working
the way I wanted and we haven't had any problems with it. I don't
particularly like the memcached backend, which looks like the only
current option (other than a database) to cluster idp 3. We currently
delegate idp authentication to CAS, and don't enable the idp backdoor
port or artifact resolution, so my current idp deployment uses stateless
clustering. Works really well so far.

 things for the IdP and they will probably emerge in the near future if
 folks need or want them. On balance, the attribute engine of the IdP is a
 powerful capability that has no analogue in Jasig CAS.

Yes, from what I've seen the attribute filter basically functions as the
service registry for CAS clients, but gives you the idp feature of only
allowing certain values of some attributes rather than all.

 I could probably come up with some more pros and cons to discuss, but the
 ones I've listed seem most notable to me as a deployer. If there are any
 features of interest to you in particular, please mention them and we can
 discuss further.

Thanks. I'm not necessarily sure what interests me in particular at this
point ;), I don't know what I don't know :). As I mentioned, we're
definitely going to upgrade to idp 3 this year, and I need to decide
whether to keep a separate CAS deployment (and upgrade it to 4.1 when
it's released) or migrate CAS clients to the idp CAS support. That
decision is really going to drive how I deploy idp 3, if auth remains
delegated to CAS it will likely be stateless clustering; otherwise, I'll
need to either set it up with a stateful backend or revert to our
previous active/passive failover idp deployment.


-- 
Paul B. Henson  |  (909) 979-6361  |  http://www.cpp.edu/~henson/
Operating Systems and Network Analyst  |  hen...@cpp.edu
California State Polytechnic University  |  Pomona CA 91768

-- 
You are currently subscribed to cas-user@lists.jasig.org as: 
arch...@mail-archive.com
To unsubscribe, change settings or access archives, see 
http://www.ja-sig.org/wiki/display/JSG/cas-user


Re: [cas-user] shib idp 3 CAS support

2015-07-12 Thread Marvin Addison

 I was just wondering if anybody here is familiar with both and might be
 able to provide some pros/cons of consolidating CAS into the idp vs
 continuing to maintain a separate CAS deployment?


I'm very familiar with both from a development and deployment perspective.
Much of the growth of Jasig CAS over the past several years has been in the
area of multi-protocol support, so you'll lose anything that's not CAS or
SAML; OAuth protocol support comes to mind. That's probably the biggest
functional area of distinction.

I would say Jasig CAS has more mature management capabilities; the service
manager UI comes to mind immediately. That said, it's only a matter of time
and effort before the IdP catches up in that area. There are also a lot
more integration options for Jasig CAS, mostly in the area of storage
backends like Ehcache and Hazelcast. Again, it's easy to develop these
things for the IdP and they will probably emerge in the near future if
folks need or want them. On balance, the attribute engine of the IdP is a
powerful capability that has no analogue in Jasig CAS.

I could probably come up with some more pros and cons to discuss, but the
ones I've listed seem most notable to me as a deployer. If there are any
features of interest to you in particular, please mention them and we can
discuss further.

Best,
M

-- 
You are currently subscribed to cas-user@lists.jasig.org as: 
arch...@mail-archive.com
To unsubscribe, change settings or access archives, see 
http://www.ja-sig.org/wiki/display/JSG/cas-user