|
Post by tophercantrell on Oct 30, 2018 13:13:50 GMT
I may be mistaken here (I am still a noob and learning), but Redfish arrays are fixed length, right? The client cannot patch back a shorter or longer list for the array. For true lists that grow/shrink I would use a collection, right?
If my implementation of EthernetInterface.StaticNameServers supports two entries, then back and forth I would always send two entries.
Most other arrays have two types: the "real" type and "null" to use for "empty slots".
If this were the case for EthernetInterface.StaticNameServers then we could use entries like ["1.2.3.4",null] and [null,"1.2.3.4"] and [null,null].
Since "null" is not an option, should we use an empty string like ["1.2.3.4",""] and ["","1.2.3.4"] and ["",""] ?
|
|
|
Post by tophercantrell on Jul 12, 2018 14:25:46 GMT
The current schema (LogEntry.v1_3_0.json) includes only three event severities: OK, Warning, and Critical.
That gives us just one "everything is ok". But even "OK" messages have differing importance.
Using SYSLOG, for instance, a user logging in might be a NOTICE (Normal but significant). As opposed to a user fat-fingering their password (invalid login attempt) which might only be INFORMATIONAL. Five consecutive login failures on the same session -- likely a WARNING. If the login code reports it fell back on LDAP to check the credentials -- that's probably just DEBUG. If the Line Printer is on fire -- let's get an ALERT.
Any chance of getting additional severities added to this enum so we can better filter/route events?
|
|
|
Post by tophercantrell on Jul 9, 2018 14:57:13 GMT
Our Redfish implementation has no synchronization between callers. For example, one caller sends a POST action to clear the logs. This takes a second or two to clear. The caller doesn't get a response back until it is done.
But while the first caller is waiting, another request comes in for the same resource. The second caller gets the state of the logs halfway through being cleared.
Is there an agreed upon mechanism to have one caller wait on a resource that is being changed?
Thanks for the assist!
|
|
|
Post by tophercantrell on Jun 25, 2018 21:20:32 GMT
The schema for EthernetInterfaces.v1_4_0->StatelessAddressAutoConfiguration includes a boolean for IPv4AutoConfigEnabled and IPv6AutoConfigEnabled. redfish.dmtf.org/schemas/v1/EthernetInterface.v1_4_0.jsonThe description of the boolean is: description: "Indicates whether IPv4 SLAAC is enabled for this interface." Isn't SLAAC an IPv6-only thing? (My coworker, Brad Lanford asked me about this one. We are out of our elements here on this networking stuff ... I more than he.) Thanks for your awesome support!
|
|
|
Post by tophercantrell on May 29, 2018 20:51:48 GMT
That's a good idea ... and a good catch on the potential problem! Let me mull on that.
|
|
|
Post by tophercantrell on May 29, 2018 14:09:51 GMT
Our product has multiple network interface connections. We would like to configure protocols like SSDP and SSH on a per-interface basis.
Some of our NICs need SSH disabled. Some NICs need to have SSDP disabled. Some interfaces might need a different SSDP hop count or IPV6 scope.
Redfish models the interfaces in the "Managers/<id>/EthernetInterfaces" collection. Here we can configure DHCP and static addresses on the individual NICs.
Redfish models the protocols like SSDP and SSH in the "Managers/<id>/NetworkProtocol" object. These settings appear to be global -- not per-interface.
Is there a preferred way to model per-interface protocol settings?
|
|
|
Post by tophercantrell on Apr 25, 2018 15:25:19 GMT
Are empty strings common in Redfish payloads? I don't see anything in the specification that prohibits empty strings for things like "Manufacturer" and "Name", though maybe "null" is a better value in those cases. Very specifically, should our UI give the user the ability to "blank" a text field and send an empty string? I am trying to think of good examples where that would be needed. Thanks for the assist!
|
|
|
Post by tophercantrell on Mar 5, 2018 21:25:46 GMT
Sorry for the noob question. Our product has a number of USB ports. We are wondering where (or if) we should show a collection of USB ports. Should it be an OEM? Or should we use the generic Switch/Port schemas?
|
|
|
Post by tophercantrell on Feb 8, 2018 14:29:29 GMT
Thank you so much!
|
|
|
Post by tophercantrell on Feb 6, 2018 14:53:34 GMT
Thank you so much! You are providing great service!
|
|
|
Post by tophercantrell on Feb 6, 2018 14:47:23 GMT
The Chassis.Power.PowerSupplies attribute is an array. I am browsing the Redfish mockups. Each item in the array is an object that has an "@odata.id". The last part of the id is "/PowerSupplies/0" or "/PowerSupplies/1" or "/PowerSupplies/N" where N is the index in the array.
Do all Redfish array items use the index like this on the end to identify a single element in the array?
In each entry in the PowerSupplies markup there is a MemberId attribute that follows the index. But the schema says that MemberId is optional. That leads me to believe the odata.id should end in the index.
Am I thinking correctly?
|
|
|
Post by tophercantrell on Feb 2, 2018 15:28:59 GMT
FunctionClasses in Memory.v1_3_0 is type "array". The "items" definition has a "deprecated" tag in it. But it is the outer "FunctionClasses" that has been deprecated -- not the "item" tag itself. This is the only place in the schema where "items" has a "deprecated" tag. Is this a typo?
|
|
|
Post by tophercantrell on Feb 2, 2018 15:02:14 GMT
Many of the Manager.json versions e.g. Manager.v1_0_5.json have a definition for "ManagerService". The definition is always empty and never referenced anywhere else in the schema. Is this for future expansion?
|
|
|
Post by tophercantrell on Jul 7, 2017 21:18:48 GMT
JSON does not different numeric types like "integer" and "float". Everything is just a "number", which can be integers or floats.
Some values in the schema like "PowerConsumedWatts" are definitely floats. Some values in the schema like "MaxPasswordLength" are definitely integers.
The schema definition defines both values as simply "number". Is there an easy way (other than with context) for clients to know how to restrict user inputs to pure integers where appropriate?
|
|
|
Post by tophercantrell on May 30, 2017 12:16:41 GMT
That all makes perfect sense. Thanks for detailed information and example!
|
|